Yves Robert

Mapping Linear Workflows with Computation/Communication Overlap 1 (2009)

Kunal Agrawal, Anne Benoit, Yves Robert

This paper presents theoretical results related to mapping and scheduling linear workflows onto heterogeneous platforms. We use a realistic architectural model with bounded communication capabilities...

Optimizing the Latency of Streaming Applications under Throughput and Reliability Constraints (2009)

Benoit, Anne, Hakem, Mourad, Robert, Yves

In this paper, we deal with the problem of scheduling streaming applications on unreliable heterogeneous platforms. We use the realistic one-port model with full computation/communication overlap. We...

Optimizing the Latency of Streaming Applications under Throughput and Reliability Constraints (2009)

Benoit, Anne, Hakem, Mourad, Robert, Yves

In this paper, we deal with the problem of scheduling streaming applications on unreliable heterogeneous platforms. We use the realistic one-port model with full computation/communication overlap. We...

Complexity analysis and performance evaluation of matrix product on multicore architectures (2009)

Jacquelin, Mathias, Marchal, Loris, Robert, Yves

The multicore revolution is underway, bringing new chips introducing more complex memory architectures. Classical algorithms must be revisited in order to take the hierarchical memory layout into...

Complexity analysis and performance evaluation of matrix product on multicore architectures (2009)

Jacquelin, Mathias, Marchal, Loris, Robert, Yves

The multicore revolution is underway, bringing new chips introducing more complex memory architectures. Classical algorithms must be revisited in order to take the hierarchical memory layout into...

Mapping Filtering Streaming Applications With Communication Costs (2009)

Agrawal, Kunal, Benoit, Anne, Dufossé, Fanny, Robert, Yves

In this paper, we explore the problem of mapping filtering streaming applications on large-scale homogeneous platforms, with a particular emphasis on communication models and their impact. Filtering...

Mapping Filtering Streaming Applications With Communication Costs (2009)

Agrawal, Kunal, Benoit, Anne, Dufossé, Fanny, Robert, Yves

In this paper, we explore the problem of mapping filtering streaming applications on large-scale homogeneous platforms, with a particular emphasis on communication models and their impact. Filtering...

Static Worksharing Strategies for Heterogeneous Computers with Unrecoverable Failures (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is "divisible" (its constituent work's granularity can be adjusted arbitrarily) and one has access to p remote computers that can assist in computing the workload. How...

Static Worksharing Strategies for Heterogeneous Computers with Unrecoverable Failures (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is "divisible" (its constituent work's granularity can be adjusted arbitrarily) and one has access to p remote computers that can assist in computing the workload. How...

Static Strategies for Worksharing with Unrecoverable Interruptions (Extended version) (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is ``divisible''---its constituent work's granularity can be adjusted arbitrarily---and one has access to p remote computers that can assist in computing the workload....

Static Strategies for Worksharing with Unrecoverable Interruptions (Extended version) (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is ``divisible''---its constituent work's granularity can be adjusted arbitrarily---and one has access to p remote computers that can assist in computing the workload....

Static Strategies for Worksharing with Unrecoverable Interruptions (Extended version) (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is "divisible''---its constituent work's granularity can be adjusted arbitrarily---and one has access to $p$ remote computers that can assist in computing the workload....

Static Strategies for Worksharing with Unrecoverable Interruptions (Extended version) (2009)

Benoit, Anne, Robert, Yves, Rosenberg, Arnold, Vivien, Frédéric

One has a large workload that is "divisible''---its constituent work's granularity can be adjusted arbitrarily---and one has access to $p$ remote computers that can assist in computing the workload....

Resource allocation strategies for constructive in-network stream processing (2009)

Benoit, Anne, Casanova, Henri, Rehn, Veronika, Robert, Yves

We consider the operator mapping problem for in-network stream processing, i.e., the application of a tree of operators in steady-state to multiple data objects that are continuously updated at...

Resource allocation strategies for constructive in-network stream processing (2009)

Benoit, Anne, Casanova, Henri, Rehn, Veronika, Robert, Yves

We consider the operator mapping problem for in-network stream processing, i.e., the application of a tree of operators in steady-state to multiple data objects that are continuously updated at...

Resource Allocation for Multiple Concurrent In-network Stream-processing Applications (2009)

Benoit, Anne, Casanova, Henri, Rehn, Veronika, Robert, Yves

Thispaperinvestigatestheoperatormappingproblemforin- network stream-processing applications. In-network stream-processing is the application of one or several trees of operators, in steady-state, to...

Resource Allocation for Multiple Concurrent In-network Stream-processing Applications (2009)

Benoit, Anne, Casanova, Henri, Rehn, Veronika, Robert, Yves

Thispaperinvestigatestheoperatormappingproblemforin- network stream-processing applications. In-network stream-processing is the application of one or several trees of operators, in steady-state, to...

Tiling for heterogeneous computing platforms (2008)

Pierre Boulet, Jack Dongarra, Yves Robert, Frederic Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a source-to-source program transformation. However, little work has been devoted to the mapping and scheduling of...

Evolution of parallel machines (2008)

Yves Robert, École Normale, Supérieure Lyon, Larry Carter, Henri Casanova, Jeanne Ferrante, ...

Evolution of parallel machines... and soon to the Holy Grid? Parallel algorithm design and scheduling were already difficult tasks with homogeneous machines Yves Robert Scheduling for Heterogeneous...

COMPUTING APPLICATIONS STATIC LU DECOMPOSITION STATIC LU DECOMPOSITION ON HETEROGENEOUS PLATFORMS (2008)

Olivier Beaumont, Arnaud Legr, Fabrice Rastello, Yves Robert, Olivier Beaumont, Arnaud Legrand, ...

In this paper, the authors deal with algorithmic issues on heterogeneous platforms. They concentrate on dense linear algebra kernels, such as matrix multiplication or LU decomposition. Block-cyclic...

References Books and edited proceedings (2008)

Yves Robert

[2] Y. Robert. The impact of vector and parallel architectures on the Gaussian elimination algorithm. Manchester University Press and John Wiley, 1991. [3] L. Bougé, C. Kenyon, J-M. Muller, and Y....

Laboratoire de l’Informatique du Parallélisme (2008)

École Normale, Supérieure Lyon, Unité Mixte, Olivier Beaumont, Loris Marchal, ...

Scheduling divisible loads with return messages on heterogeneous master-worker platforms

its Applications, 48:105–117, 1982. (2008)

Yves Robert

[1] Yves Robert. Regular incomplete factorizations of real positive definite matrices. Linear Algebra and

Mapping and load-balancing iterative computations on heterogeneous clusters (2008)

Arnaud Legrand, Hélène Renard, Yves Robert, Frédéric Vivien

This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each iteration,...

Scheduling multiple divisible loads on a linear processor network (2008)

Matthieu Gallet, Yves Robert, Frédéric Vivien, Thème Num, Matthieu Gallet, Yves Robert, ...

apport de recherche ISSN 0249-6399 ISRN INRIA/RR--6235--FR+ENG Scheduling multiple divisible loads on a linear processor network

References Conferences (2008)

Yves Robert

[3] Yves Robert and Maurice Tchuente. Special-purpose architectures for string processing. In COGNI-

International Journal of Foundations of Computer Science c ○ World Scientific Publishing Company REVISITING MATRIX PRODUCT ON MASTER-WORKER PLATFORMS (2008)

Jack Dongarra, Jean-françois Pineau, Yves Robert, Zhiao Shi, Frédéric Vivien

Communicated by Editor’s name This paper is aimed at designing efficient parallel matrix-product algorithms for homogeneous master-worker platforms. While matrix-product is well-understood for...

Mapping and Load-Balancing Iterative Computations on Heterogeneous Clusters (2008)

Arnaud Legr, Hélène Renard, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each...

The`me NUM (2008)

Cois Pineau, Yves Robert

The impact of heterogeneityon master-slave on-line scheduling

The`me NUM (2008)

Loris Marchal, Veronika Rehn, Yves Robert, Loris Marchal, Veronika Rehn, Yves Robert, ...

Scheduling and data redistribution strategies on star platforms

Data redistribution algorithms for homogeneous and heterogeneous processor rings (2008)

Hélène Renard, Yves Robert, Frédéric Vivien

Abstract. We consider the problem of redistributing data on homogeneous and heterogeneous processor rings. The problem arises in several applications, each time after a load-balancing mechanism is...

Bi-criteria Pipeline Mappings for Parallel Image Processing (2008)

École Normale, Supérieure Lyon, Anne Benoit, Harald Kosch, Veronika Rehn-sonigo, Yves Robert, ...

Mapping workflow applications onto parallel platforms is a challenging problem, even for simple application patterns such as pipeline graphs. Several antagonistic criteria should be optimized, such...

Realistic Models and Efficient Algorithms for Fault Tolerant Scheduling on Heterogeneous Platforms (2008)

École Normale, Supérieure Lyon, Anne Benoit, Mourad Hakem, Yves Robert, École Normale, ...

Most list scheduling heuristics rely on a simple platform model where communication contention is not taken into account. In addition, it is generally assumed that processors in the systems are...

Mapping linear workflows with computation/communication overlap (2008)

Kunal Agrawal, Anne Benoit, Yves Robert

This paper presents theoretical results related to mapping and scheduling linear workflows onto heterogeneous platforms. We use a realistic architectural model with bounded communication capabilities...

On the Removal of Anti and Output Dependences (2007)

Frdric Vivien, Pierre-yves Call, Alain Darte, Alain Darte, Yves Robert, ...

: In this paper we build upon results of Padua and Wolfe [8], who introduce two graph transformations to eliminate anti and output dependences. We first give a unified framework for such...

Technology Transfer within the ProHPC TTN at ENS Lyon (2007)

Christophe Barberet, Lionell Brunie, Frederic Desprez, Gilles Lebourgeois, Raymond Namyst, Yves Robert, ...

. This article is devoted to the description of our activities related to transferring the HPCN technology to SMEs. This work is performed in the framework of the French TTN ProHPC which we briey...

More on Scheduling Block-Cyclic Array Redistribution (2007)

Frédéric Desprez, Stéphane Domas, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper [2], we have reported how to...

PVM implementation of heterogeneous ScaLAPACK dense linear solvers (Extended Abstract) (2007)

Vincent Boudet, Fabrice Rastello, Yves Robert

Vincent Boudet, Fabrice Rastello and Yves Robert LIP, UMR CNRS{ENS Lyon{INRIA 5668 Ecole Normale Superieure de Lyon F - 69364 Lyon Cedex 07 e-mail: Vincent.Boudet@ens-lyon.fr Abstract. We report only...

Retiming DAGs (2007)

Retiming Dags, P.Y. Calland, A. Mignotte, O. Peyran, Y. Robert, F. Vivien, ...

This paper is devoted to a low-complexity algorithm for retiming circuits without cycles, i.e. whose network graph is a Direct Acyclic Graph (DAG). On one hand DAGs have a great practical importance,...

Extended (2007)

Nd Ed, Unite Mixte, Vincent Boudet, Vincent Boudet, Fabrice Rastello, ...

Vincent Boudet, Fabrice Rastello and Yves Robert March 1999 Research Report N o 1999-19 Ecole Normale Sup erieure de Lyon 46 Allee d'Italie, 69364 Lyon Cedex 07, France Telephone :...

CS 594 - Understanding Parallel Architectures: From Theory To Practice (2007)

Jack Dongarra, Yves Robert, Distinguished Professor, Ta Caroline Papadopoulos, Rm Ayres Hall

n libraries like MPI, general-purpose task systems such as PVM, or even data-parallel languages like HPF. These are the tools that the class targets. In this course we plan to cover and understand...

More on Scheduling Block-Cyclic Array Redistribution (2007)

Fr'ed'eric Desprez, Jack Dongarra, Jack Dongarra, Antoine Petitet, Antoine Petitet, Cyril R, ...

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper, we have reported how to...

Laboratoire de l'Informatique du Paralllisme (2007)

Unit Mixte, Olivier Beaumont, Olivier Beaumont, ...

A polynomial-time algorithm for allocating independent tasks on heterogeneous fork-graphs

Laboratoire de l'Informatique du Paralllisme (2007)

Unit Mixte, Arnaud Legr, Helene Renard, ...

Load-balancing iterative computations in heterogeneous clusters with shared communication links Arnaud Legrand,

J. Parallel and Distributed Computing, 2001, paratre. (2007)

Christophe Barberet, Lionel Brunie, Frdric Desprez, Gilles Lebourgeois, Raymond Namyst, Stphane Ubeda, ...

[74] Pierre Boulet, Jack Dongarra, Yves Robert, and Frdric Vivien. Static tiling for heterogeneous computing

A Realistic Model and an Ecient Heuristic (2007)

Unit Mixte, Vincent Boudet, Olivier Beaumont, ...

Scheduling computational tasks on processors is a key issue for highperformance computing. Although a large number of scheduling heuristics have been presented in the literature, most of them target...

Arnaud Legrand, (2007)

Unit Mixte, Loris Marchal, Yves Robert, ...

Optimizing the steady-state throughput of scatter and reduce operations on heterogeneous platforms

Algorithmic Issues on Heterogeneous Computing Platforms (2007)

Unit Mixte, Fr#d#ric Vivien, Pierre Boulet, ...

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). Dealing with processors of dioeerent speeds...

Loop Partitioning for Cache-based Multiprocessors (2007)

Fabrice Rastello, Yves Robert

In this paper, an ecient algorithm to implement loop partitioning is introduced and evaluated. We improve recent results of Agarwal, Kranz and Natarajan [1] in several directions. We derive a new...

Scheduling tasks sharing les on heterogeneous clusters (2007)

Frederic Vivien, Arnaud Giersch, Arnaud Giersch, Yves Robert, Yves Robert, ...

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) les which initially reside on a master processor. A given le may...

Algorithmic Issues on Heterogeneous Computing Platforms (2007)

Unite Mixte, Ecole Normale, Sup Lyon, Pierre Boulet, Pierre Boulet, ...

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). Dealing with processors of dierent speeds...

Laboratoire de l'Informatique du Paralllisme (2007)

Unit Mixte, Arnaud Legr, Helene Renard, ...

Load-balancing iterative computations in heterogeneous clusters with shared communication links Arnaud Legrand,

4 (2007)

Pierre Boulet, Jack Dongarra, Yves Robert, Frederic Vivien

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). Dealing with processors of dierent speeds...

1 (2007)

Arnaud Giersch, Yves Robert

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) les which initially reside on a master processor. A given le may...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Comments on ``Design and performance evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks'' (2007)

Gallet, Matthieu, Robert, Yves, Vivien, Frédéric

Min, Veeravalli, and Barlas proposed strategies to minimize the overall execution time of one or several divisible loads on a heterogeneous linear network, using one or more installments. We show on...

Comments on ``Design and performance evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks'' (2007)

Gallet, Matthieu, Robert, Yves, Vivien, Frédéric

Min, Veeravalli, and Barlas proposed strategies to minimize the overall execution time of one or several divisible loads on a heterogeneous linear network, using one or more installments. We show on...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Mapping pipeline skeletons onto heterogeneous platforms (2007)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Comments on ``Design and performance evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks'' (2007)

Gallet, Matthieu, Robert, Yves, Vivien, Frédéric

Min, Veeravalli, and Barlas proposed strategies to minimize the overall execution time of one or several divisible loads on a heterogeneous linear network, using one or more installments. We show on...

Comments on ``Design and performance evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks'' (2007)

Gallet, Matthieu, Robert, Yves, Vivien, Frédéric

Min, Veeravalli, and Barlas proposed strategies to minimize the overall execution time of one or several divisible loads on a heterogeneous linear network, using one or more installments. We show on...

Statistical Algorithms Reference Guide. http://www.affymetrix. com/support/technical/technotes/statistical_reference_guide.pdf (2007)

Anne Benoit, Yves Robert

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous –nowadays a standard assumption. A high-level approach to...

replicated and (2007)

Anne Benoit, Yves Robert

results for throughput and latency optimization of

Statistical Algorithms Reference Guide. http://www.affymetrix. com/support/technical/technotes/statistical_reference_guide.pdf (2007)

Anne Benoit, Yves Robert

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous –nowadays a standard assumption. A high-level approach to...

Fault Tolerant Scheduling of Precedence Task Graphs on Heterogeneous Platforms (2007)

Anne Benoit, Mourad Hakem, Yves Robert

Fault tolerance and latency are important requirements in several applications which are time critical in nature: such applications require guaranties in terms of latency, even when processors are...

Laboratoire de l’Informatique du Parallélisme (2007)

École Normale, Supérieure Lyon, Matthieu Gallet, Yves Robert, Frédéric Vivien, École Normale, ...

evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks”

Laboratoire de l’Informatique du Parallélisme (2007)

École Normale, Supérieure Lyon, Matthieu Gallet, Yves Robert, Frédéric Vivien, École Normale, ...

evaluation of load distribution strategies for multiple loads on heterogeneous linear daisy chain networks”

Optimizing Latency and Reliability of Pipeline Workflow Applications (2007)

École Normale, Supérieure Lyon, Anne Benoit, Veronika Rehn-sonigo, Yves Robert, École Normale, ...

Mapping applications onto heterogeneous platforms is a difficult challenge, even for simple application patterns such as pipeline graphs. The problem is even more complex when processors are subject...

Statistical Algorithms Reference Guide. http://www.affymetrix. com/support/technical/technotes/statistical_reference_guide.pdf (2007)

Anne Benoit, Yves Robert

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous –nowadays a standard assumption. A high-level approach to...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Scheduling and data redistribution strategies on star platforms (2006)

Marchal, Loris, Rehn, Veronika, Robert, Yves, Vivien, Frédéric

In this work we are interested in the problem of scheduling and redistributing data on master-slave platforms. We consider the case were the workers possess initial loads, some of which having to be...

Scheduling and data redistribution strategies on star platforms (2006)

Marchal, Loris, Rehn, Veronika, Robert, Yves, Vivien, Frédéric

In this work we are interested in the problem of scheduling and redistributing data on master-slave platforms. We consider the case were the workers possess initial loads, some of which having to be...

Scheduling and data redistribution strategies on star platforms (2006)

Marchal, Loris, Rehn, Veronika, Robert, Yves, Vivien, Frédéric

In this work we are interested in the problem of scheduling and redistributing data on master-slave platforms. We consider the case were the workers possess initial loads, some of which having to be...

Scheduling and data redistribution strategies on star platforms (2006)

Marchal, Loris, Rehn, Veronika, Robert, Yves, Vivien, Frédéric

In this work we are interested in the problem of scheduling and redistributing data on master-slave platforms. We consider the case were the workers possess initial loads, some of which having to be...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Strategies for Replica Placement in Tree Networks (2006)

Robert, Yves, Benoit, Anne, Rehn, Veronika

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Revisiting Matrix Product on Master-Worker Platforms (2006)

Dongarra, Jack, Pineau, Jean-François, Robert, Yves, Shi, Zhiao, Vivien, Frédéric

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of...

Revisiting Matrix Product on Master-Worker Platforms (2006)

Dongarra, Jack, Pineau, Jean-François, Robert, Yves, Shi, Zhiao, Vivien, Frédéric

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of...

Revisiting Matrix Product on Master-Worker Platforms (2006)

Dongarra, Jack, Pineau, Jean-François, Robert, Yves, Shi, Zhiao, Vivien, Frédéric

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of...

Revisiting Matrix Product on Master-Worker Platforms (2006)

Dongarra, Jack, Pineau, Jean-François, Robert, Yves, Shi, Zhiao, Vivien, Frédéric

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of...

Mapping pipeline skeletons onto heterogeneous platforms (2006)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Impact of QoS on Replica Placement in Tree Networks (2006)

Benoit, Anne, Rehn, Veronika, Robert, Yves

This paper discusses and compares several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number and...

Impact of QoS on Replica Placement in Tree Networks (2006)

Benoit, Anne, Rehn, Veronika, Robert, Yves

This paper discusses and compares several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number and...

Mapping pipeline skeletons onto heterogeneous platforms (2006)

Benoit, Anne, Robert, Yves

Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous --nowadays a standard assumption. A high-level approach to...

Impact of QoS on Replica Placement in Tree Networks (2006)

Anne Benoit, Veronika Rehn, Yves Robert

This paper discusses and compares several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number and...

Strategies for Replica Placement in Tree Networks (2006)

École Normale, Supérieure Lyon, Anne Benoit, Veronika Rehn, Yves Robert, École Normale, ...

In this paper, we discuss and compare several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number...

Revisiting matrix product on master-worker platforms (2006)

Jean-françois Pineau, Yves Robert, Frédéric Vivien, Zhiao Shi, Jack Dongarra

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of...

Centralized Versus Distributed Schedulers (2006)

Olivier Beaumont, Larry Carter, Jeanne Ferrante, Arnaud Legr, Loris Marchal, ...

Multiple applications that execute concurrently on heterogeneous platforms compete for CPU and network resources. In this paper we consider the problem of scheduling applications to ensure fair and...

Matthieu Gallet, (2006)

École Normale, Supérieure Lyon, Unité Mixte, Yves Robert, Frédéric Vivien, ...

Scheduling communication requests traversing a switch: complexity and algorithms

The impact of heterogeneity on master-slave on-line scheduling, in "HCW’2006, the 15th Heterogeneous Computing Workshop", to appear (2006)

Jean-françois Pineau, Yves Robert, Frédéric Vivien

In this paper, we assess the impact of heterogeneity for scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single...

Revisiting matrix product on master-worker platforms (2006)

Jack Dongarra, Revisiting Matrix, Product Master-worker, Platforms Jack Dongarra, Cois Pineau, Cois Pineau, ...

Laboratoire de l'Informatique du Paralle'lisme E'cole Normale Supe'rieure de LyonUnite ' Mixte de Recherche CNRS-INRIA-ENS LYON-UCBL no 5668

Revisiting matrix product on master-worker platforms (2006)

Jack Dongarra, Jean-françois Pineau, Yves Robert, Frédéric Vivien

This paper is aimed at designing efficient parallel matrix-product algorithms for homogeneous masterworker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of processors...

Revisiting Matrix Product on Master-Worker Platforms (2006)

École Normale, Supérieure Lyon, Jack Dongarra, Zhiao Shi, Frédéric Vivien, École Normale, ...

This paper is aimed at designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is wellunderstood for homogeneous 2D-arrays of processors...

Impact of QoS on Replica Placement in Tree Networks (2006)

École Normale, Supérieure Lyon, Anne Benoit, Veronika Rehn, Yves Robert, École Normale, ...

This paper discusses and compares several policies to place replicas in tree networks, subject to server capacity and QoS constraints. The client requests are known beforehand, while the number and...

FIFO scheduling of divisible loads with return messages under the one-port model (2005)

Beaumont, Olivier, Marchal, Loris, Rehn, Veronika, Robert, Yves

This paper deals with scheduling divisible load applications on star networks, in presence of return messages. This work is a follow-on of , where the same problem was considered under the 2-port...

Scheduling multiple bags of tasks on heterogeneous master- worker platforms: centralized versus distributed solutions (2005)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Marchal, Loris, Robert, Yves

Multiple applications that execute concurrently on heterogeneous platforms compete for CPU and network resources. In this paper we consider the problem of scheduling applications to ensure fair and...

The impact of heterogeneity on master-slave on-line scheduling (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this paper, we assess the impact of heterogeneity for scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single...

Off-line and on-line scheduling on heterogeneous master-slave platforms (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this work, we deal with the problem of scheduling independent tasks on heterogeneous master-slave platforms. We target both off-line and on-line problems, with several objective functions...

Optimizing Network Resource Sharing in Grids (2005)

Marchal, Loris, Robert, Yves, Zeng, Jingdi

While grid computing reaches further to geographically separated clusters, data warehouses, and disks, it poses demanding requirements on end-to-end performance guarantee. Its pre-defined...

FIFO scheduling of divisible loads with return messages under the one-port model (2005)

Beaumont, Olivier, Marchal, Loris, Rehn, Veronika, Robert, Yves

This paper deals with scheduling divisible load applications on star networks, in presence of return messages. This work is a follow-on of , where the same problem was considered under the 2-port...

Scheduling multiple bags of tasks on heterogeneous master- worker platforms: centralized versus distributed solutions (2005)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Marchal, Loris, Robert, Yves

Multiple applications that execute concurrently on heterogeneous platforms compete for CPU and network resources. In this paper we consider the problem of scheduling applications to ensure fair and...

The impact of heterogeneity on master-slave on-line scheduling (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this paper, we assess the impact of heterogeneity for scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single...

Off-line and on-line scheduling on heterogeneous master-slave platforms (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this work, we deal with the problem of scheduling independent tasks on heterogeneous master-slave platforms. We target both off-line and on-line problems, with several objective functions...

Optimizing Network Resource Sharing in Grids (2005)

Marchal, Loris, Robert, Yves, Zeng, Jingdi

While grid computing reaches further to geographically separated clusters, data warehouses, and disks, it poses demanding requirements on end-to-end performance guarantee. Its pre-defined...

FIFO scheduling of divisible loads with return messages under the one-port model (2005)

Beaumont, Olivier, Marchal, Loris, Rehn, Veronika, Robert, Yves

This paper deals with scheduling divisible load applications on star networks, in presence of return messages. This work is a follow-on of , where the same problem was considered under the 2-port...

Scheduling multiple bags of tasks on heterogeneous master- worker platforms: centralized versus distributed solutions (2005)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Marchal, Loris, Robert, Yves

Multiple applications that execute concurrently on heterogeneous platforms compete for CPU and network resources. In this paper we consider the problem of scheduling applications to ensure fair and...

The impact of heterogeneity on master-slave on-line scheduling (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this paper, we assess the impact of heterogeneity for scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single...

Off-line and on-line scheduling on heterogeneous master-slave platforms (2005)

Pineau, Jean-François, Robert, Yves, Vivien, Frédéric

In this work, we deal with the problem of scheduling independent tasks on heterogeneous master-slave platforms. We target both off-line and on-line problems, with several objective functions...

Optimizing Network Resource Sharing in Grids (2005)

Marchal, Loris, Robert, Yves, Zeng, Jingdi

While grid computing reaches further to geographically separated clusters, data warehouses, and disks, it poses demanding requirements on end-to-end performance guarantee. Its pre-defined...

Olivier Beaumont, Loris Marchal, (2005)

École Normale, Supérieure Lyon, Unité Mixte, Veronika Rehn, Yves Robert, ...

FIFO scheduling of divisible loads with return messages under the one-port model

Olivier Beaumont, (2005)

École Normale, Supérieure Lyon, Unité Mixte, Larry Carter, Jeanne Ferrante, ...

Scheduling multiple bags of tasks on heterogeneous master-worker platforms: centralized versus distributed solutions

Optimizing Network Resource Sharing in Grids (2005)

École Normale, Supérieure Lyon, Loris Marchal, Yves Robert, Jingdi Zeng, ...

While grid computing reaches further to geographically separated clusters, data warehouses, and disks, it poses demanding requirements on end-to-end performance guarantee. Its pre-defined...

Broadcast trees for heterogeneous platforms (2005)

Olivier Beaumont, Loris Marchal, Yves Robert, Olivier Beaumont, Loris Marchal, Yves Robert

Laboratoire de l'Informatique du Paralle'lisme E'cole Normale Supe'rieure de LyonUnite ' Mixte de Recherche CNRS-INRIA-ENS LYON-UCBL no 5668

The`me NUM (2005)

Cois Pineau, Yves Robert

Off-line and on-line schedulingon heterogeneous master-slave platforms

The impact of heterogeneity on master-slave on-line scheduling (2005)

École Normale, Supérieure Lyon, Yves Robert, Frédéric Vivien, École Normale, Supérieure Lyon, ...

In this paper, we assess the impact of heterogeneity for scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single...

Scheduling network requests with transmission window (2005)

École Normale, Supérieure Lyon, Loris Marchal, Yves Robert, Jingdi Zeng, École Normale, ...

We consider the problem of bulk data transfers and bandwidth sharing in the context of grid infrastructures. Grid computing empowers high-performance computing in a large-scale distributed...

Independent and Divisible Task Scheduling on Heterogeneous Star-shaped Platforms with Limited Memory (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the problem of allocating and scheduling a collection of independent, equal-sized tasks on heterogeneous star-shaped platforms. We also address the same problem for...

A realistic network/application model for scheduling divisible loads on large-scale platforms (2004)

Marchal, Loris, Yang, Yang, Casanova, Henri, Robert, Yves

Divisible load applications consist of an amount of data and associated computation that can be divided arbitrarily into any number of independent pieces. This model is a good approximation of many...

Assessing the impact and limits of steady-state scheduling for mixed task and data parallelism on heterogeneous platforms (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider steady-state scheduling techniques for mapping a collection of application graphs onto heterogeneous systems, such as clusters and grids. We advocate the use of...

Data redistribution algorithms for heterogeneous processor rings (2004)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

We consider the problem of redistributing data on homogeneous and heterogeneous ring of processors. The problem arises in several applications, each time after that a load-balancing mechanism is...

Scheduling Tasks Sharing Files from Distributed Repositories (revised version) (2004)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Independent and Divisible Task Scheduling on Heterogeneous Star-shaped Platforms with Limited Memory (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the problem of allocating and scheduling a collection of independent, equal-sized tasks on heterogeneous star-shaped platforms. We also address the same problem for...

A realistic network/application model for scheduling divisible loads on large-scale platforms (2004)

Marchal, Loris, Yang, Yang, Casanova, Henri, Robert, Yves

Divisible load applications consist of an amount of data and associated computation that can be divided arbitrarily into any number of independent pieces. This model is a good approximation of many...

Assessing the impact and limits of steady-state scheduling for mixed task and data parallelism on heterogeneous platforms (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider steady-state scheduling techniques for mapping a collection of application graphs onto heterogeneous systems, such as clusters and grids. We advocate the use of...

Data redistribution algorithms for heterogeneous processor rings (2004)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

We consider the problem of redistributing data on homogeneous and heterogeneous ring of processors. The problem arises in several applications, each time after that a load-balancing mechanism is...

Complexity results and heuristics for pipelined multicast operations on heterogeneous platforms (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application deployed on a heterogeneous platform. Such applications extensively use macro-communication schemes,...

Scheduling Tasks Sharing Files from Distributed Repositories (revised version) (2004)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Independent and Divisible Task Scheduling on Heterogeneous Star-shaped Platforms with Limited Memory (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the problem of allocating and scheduling a collection of independent, equal-sized tasks on heterogeneous star-shaped platforms. We also address the same problem for...

A realistic network/application model for scheduling divisible loads on large-scale platforms (2004)

Marchal, Loris, Yang, Yang, Casanova, Henri, Robert, Yves

Divisible load applications consist of an amount of data and associated computation that can be divided arbitrarily into any number of independent pieces. This model is a good approximation of many...

Assessing the impact and limits of steady-state scheduling for mixed task and data parallelism on heterogeneous platforms (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider steady-state scheduling techniques for mapping a collection of application graphs onto heterogeneous systems, such as clusters and grids. We advocate the use of...

Data redistribution algorithms for heterogeneous processor rings (2004)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

We consider the problem of redistributing data on homogeneous and heterogeneous ring of processors. The problem arises in several applications, each time after that a load-balancing mechanism is...

Complexity results and heuristics for pipelined multicast operations on heterogeneous platforms (2004)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application deployed on a heterogeneous platform. Such applications extensively use macro-communication schemes,...

Scheduling Tasks Sharing Files from Distributed Repositories (revised version) (2004)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Scheduling Tasks Sharing Files from Distributed Repositories (2004)

Arnaud Giersch, Frederic Vivien, Arnaud Giersch, Yves Robert, Yves Robert

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Data Redistribution Algorithms for Homogeneous and Heterogeneous Processor Rings (2004)

Helene Renard, Yves Robert, Frederic Vivien

We consider the problem of redistributing data on homogeneous and heterogeneous processor rings. The problem arises in several applications, each time after a load-balancing mechanism is invoked (but...

Data Redistribution Algorithms For Heterogeneous Processor Rings (2004)

Helene Renard, Yves Robert, Frederic Vivien

We consider the problem of redistributing data on homogeneous and heterogeneous ring of processors. The problem arises in several applications, each time after that a load-balancing mechanism is...

Data redistribution algorithms for heterogeneous processor rings (2004)

École Normale, Supérieure Lyon, Hélène Renard, Yves Robert, Frédéric Vivien, École Normale, ...

We consider the problem of redistributing data on homogeneous and heterogeneous ring of processors. The problem arises in several applications, each time after that a load-balancing mechanism is...

Olivier Beaumont, Arnaud Legrand, (2004)

École Normale, Supérieure Lyon, Unité Mixte, Loris Marchal, Yves Robert, ...

Assessing the impact and limits of steady-state scheduling for mixed task and data parallelism on heterogeneous platforms

Steady-State Scheduling on Heterogeneous Clusters: Why and How? (2004)

École Normale, Supérieure Lyon, Olivier Beaumont, Arnaud Legrand, Loris Marchal, Yves Robert, ...

In this paper, we consider steady-state scheduling techniques for heterogeneous systems, such as clusters and grids. We advocate the use of steady-state scheduling to solve a variety of important...

Broadcast Trees for Heterogeneous Platforms (2004)

École Normale, Supérieure Lyon, Olivier Beaumont, Loris Marchal, Yves Robert, École Normale, ...

In this paper, we deal with broadcasting on heterogeneous platforms. Typically, the message to be broadcast is split into several slices, which are sent by the source processor in a pipeline fashion....

Independent and Divisible Task Scheduling on Heterogeneous Star-shaped Platforms with Limited Memory (2004)

École Normale, Supérieure Lyon, Olivier Beaumont, Arnaud Legrand, Loris Marchal, Yves Robert, ...

In this paper, we consider the problem of allocating and scheduling a collection of independent, equal-sized tasks on heterogeneous starshaped platforms. We also address the same problem for...

Loris Marchal, Yang Yang, (2004)

École Normale, Supérieure Lyon, Unité Mixte, Henri Casanova, Yves Robert, ...

A realistic network/application model for scheduling divisible loads on large-scale platforms

Data redistribution algorithms for heterogeneous processor rings (2004)

Yves Robert, Yves Robert

Data redistribution algorithms for heterogeneous processor rings

Scheduling tasks sharing files from distributed repositories (2004)

Arnaud Giersch, Yves Robert, Frdric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto a distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor...

Mapping and Load-Balancing Iterative Computations (2004)

Arnaud Legr, Hélène Renard, Yves Robert, Senior Member, Frédéric Vivien

Abstract—This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each...

Scheduling Tasks Sharing Files from Distributed Repositories (revised version) (2004)

École Normale, Supérieure Lyon, Arnaud Giersch, Yves Robert, Frédéric Vivien, École Normale, ...

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Scheduling tasks sharing files on heterogeneous master-slave platforms (2004)

Arnaud Giersch, Yves Robert, Frédéric Vivien

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Scheduling tasks sharing files from distributed repositories (2004)

Arnaud Giersch, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto a distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor...

Data redistribution algorithms for heterogeneous processor rings (2004)

Hélène Renard, Hélène Renard, Yves Robert, Yves Robert, Frédéric Vivien, Frédéric Vivien

We consider the problem of redistributing data on homogeneous and heterogeneous rings of processors. The problem arises in several applications, after each invocation of a load-balancing mechanism...

Scheduling tasks sharing files from distributed repositories (2004)

Arnaud Giersch, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto a distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor...

Data redistribution algorithms for heterogeneous processor rings (2004)

Thème Num, Hélène Renard, Hélène Renard, Yves Robert, Yves Robert, Frédéric Vivien, ...

apport de recherche Data redistribution algorithms for heterogeneous processor rings

Scheduling tasks sharing files from distributed repositories (2004)

Arnaud Giersch, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto a distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor...

Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Platforms (2004)

Cyril Banino, Olivier Beaumont, Larry Carter, Jeanne Ferrante, Senior Member, Arnaud Legrand, ...

In this paper, we consider the problem of allocating a large number of independent, equal-sized tasks to a heterogeneous computing platform. We use a nonoriented graph to model the platform, where...

Olivier Beaumont, Arnaud Legrand, (2004)

École Normale, Supérieure Lyon, Unité Mixte, Loris Marchal, Yves Robert, ...

Complexity results and heuristics for pipelined multicast operations on heterogeneous platforms

Scheduling tasks sharing files from distributed repositories (2003)

Giersch, Arnaud, Robert, Yves, Frédéric, Vivien

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems (2003)

Beaumount, Olivier, Casanova, Henri, Legrand, Arnaud, Robert, Yves, Yang, Yves

Applications in many scientific and engineering domains are structured in large numbers of independent tasks with low granularity. These applications can thus be naturally parallelized, typically in...

Steady-state scheduling of task graphs on heterogeneous computing platforms (2003)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent problems to be...

Optimizing the steady-state throughput of Broadcasts on heterogeneous platforms (2003)

Legrand, Arnaud, Beaumont, Olivier, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications extensively use macro-communication...

Optimizing the steady-state throughput of scatter and reduce operationson heterogeneous platforms (2003)

Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications intensively use collective...

Scheduling tasks sharing fileson heterogeneous clusters (2003)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Load-balancing iterative computations in heterogeneous clusters with shared communication links (2003)

Legrand, Arnaud, Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each iteration,...

Static load-balancing techniques for iterative computations on heterogeneous clusters (2003)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to static load balancing techniques for mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors. At each iteration,...

Scheduling tasks sharing files from distributed repositories (2003)

Giersch, Arnaud, Robert, Yves, Frédéric, Vivien

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems (2003)

Beaumont, Olivier, Casanova, Henri, Legrand, Arnaud, Robert, Yves, Yang, Yang

Applications in many scientific and engineering domains are structured in large numbers of independent tasks with low granularity. These applications can thus be naturally parallelized, typically in...

Steady-state scheduling of task graphs on heterogeneous computing platforms (2003)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent problems to be...

Optimizing the steady-state throughput of Broadcasts on heterogeneous platforms (2003)

Legrand, Arnaud, Beaumont, Olivier, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications extensively use macro-communication...

Optimizing the steady-state throughput of scatter and reduce operationson heterogeneous platforms (2003)

Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications intensively use collective...

Scheduling tasks sharing fileson heterogeneous clusters (2003)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Load-balancing iterative computations in heterogeneous clusters with shared communication links (2003)

Legrand, Arnaud, Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each iteration,...

Static load-balancing techniques for iterative computations on heterogeneous clusters (2003)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to static load balancing techniques for mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors. At each iteration,...

Scheduling tasks sharing files from distributed repositories (2003)

Giersch, Arnaud, Robert, Yves, Frédéric, Vivien

This paper is devoted to scheduling a large collection of independent tasks onto a large distributed heterogeneous platform, which is composed of a set of servers. Each server is a processor cluster...

Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems (2003)

Beaumont, Olivier, Casanova, Henri, Legrand, Arnaud, Robert, Yves, Yang, Yang

Applications in many scientific and engineering domains are structured in large numbers of independent tasks with low granularity. These applications can thus be naturally parallelized, typically in...

Steady-state scheduling of task graphs on heterogeneous computing platforms (2003)

Beaumont, Olivier, Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent problems to be...

Optimizing the steady-state throughput of Broadcasts on heterogeneous platforms (2003)

Legrand, Arnaud, Beaumont, Olivier, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications extensively use macro-communication...

Optimizing the steady-state throughput of scatter and reduce operationson heterogeneous platforms (2003)

Legrand, Arnaud, Marchal, Loris, Robert, Yves

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous «grid» platform. Such applications intensively use collective...

Scheduling tasks sharing fileson heterogeneous clusters (2003)

Giersch, Arnaud, Robert, Yves, Vivien, Frédéric

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Load-balancing iterative computations in heterogeneous clusters with shared communication links (2003)

Legrand, Arnaud, Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each iteration,...

Static load-balancing techniques for iterative computations on heterogeneous clusters (2003)

Renard, Hélène, Robert, Yves, Vivien, Frédéric

This paper is devoted to static load balancing techniques for mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors. At each iteration,...

Scheduling Tasks Sharing Files On Heterogeneous Clusters (2003)

Unit Mixte, Frederic Vivien, Arnaud Giersch, ...

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Arnaud Legrand, (2003)

École Normale, Supérieure Lyon, Unité Mixte, Hélène Renard, Yves Robert, ...

Load-balancing iterative computations in heterogeneous clusters with shared communication links

Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems (2003)

École Normale, Supérieure Lyon, Olivier Beaumount, Henri Casanova, Arnaud Legrand, Yves Robert, ...

Many applications in scientific and engineering domains are structured as large numbers of independent tasks with low granularity. These applications are thus amenable to straightforward...

Laboratoire de l’Informatique du Parallélisme (2003)

École Normale, Supérieure Lyon, Unité Mixte, Olivier Beaumont, Arnaud Legrand, ...

Steady-state scheduling of task graphs on heterogeneous computing platforms

optimizing the steady-state throughput of Broadcasts on heterogeneous platforms (2003)

Unit Mixte, Arnaud Legrand, Arnaud Legr, ...

In this paper, we consider the communications involved by the execution of a complex application, deployed on a heterogeneous "grid" platform. Such applications extensively use...

Steady-State Scheduling Of Task Graphs On Heterogeneous Computing Platforms (2003)

Unit Mixte, Arnaud Legrand, Olivier Beaumont, ...

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent...

Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems (2003)

Unit Mixte, Arnaud Legrand, Olivier Beaumount, ...

Many applications in scientific and engineering domains are structured as large numbers of independent tasks with low granularity. These applications are thus amenable to straightforward...

Load-Balancing Iterative Computations In Heterogeneous Clusters With . . . (2003)

Arnaud Legr, Helene Renard, Helene Renard, Yves Robert, Yves Robert, ...

This paper is devoted to mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors, which are arranged along a virtual ring. At each iteration,...

Scheduling tasks sharing files on heterogeneous clusters (2003)

Arnaud Giersch, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A...

Static load-balancing techniques for iterative computations on heterogeneous clusters (2003)

Hélène Renard, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to static load balancing techniques for mapping iterative algorithms onto heterogeneous clusters. The application data is partitioned over the processors. At each...

Hélène Renard, (2003)

École Normale, Supérieure Lyon, Unité Mixte, Yves Robert, Frédéric Vivien, ...

Static load-balancing techniques for iterative computations on heterogeneous clusters

Scheduling tasks sharing files on heterogeneous clusters (2003)

Arnaud Giersch, Yves Robert, Frédéric Vivien

Abstract. This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A...

Scheduling Tasks Sharing Files on Heterogeneous Clusters (2003)

Arnaud Giersch, Yves Robert, Frederic Vivien

This paper is devoted to scheduling a large collection of independent tasks onto heterogeneous clusters. The tasks depend upon (input) files which initially reside on a master processor. A given file...

Hélène Renard, (2003)

École Normale, Supérieure Lyon, Unité Mixte, Yves Robert, Frédéric Vivien, ...

Static load-balancing techniques for iterative computations on heterogeneous clusters

Optimal algorithms for scheduling divisible workloads on heterogeneous systems (2002)

Beaumont, Olivier, Robert, Yves

In this paper, we discuss several algorithms for scheduling divisible loads on heterogeneous systems. Our main contributions are (i) new optimality results for single-round algorithms and (ii) the...

Optimal algorithms for scheduling divisible workloads on heterogeneous systems (2002)

Beaumont, Olivier, Robert, Yves

In this paper, we discuss several algorithms for scheduling divisible loads on heterogeneous systems. Our main contributions are (i) new optimality results for single-round algorithms and (ii) the...

Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Grids (2002)

Unit Mixte, Cyril Banino, Olivier Beaumont, Arnaud Legrand, ...

In this paper, we consider the problem of allocating a large number of independent, equal-sized tasks to a heterogeneous "grid" computing platform. We use a non-oriented graph to model a...

Automatic Partitioning of Parallel Loops with Parallelepiped-Shaped Tiles (2002)

Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to implement loop partitioning is introduced and evaluated. We start from results of Agarwal et al. [1] whose aim is to minimize the number of accessed data...

Scheduling Strategies for Mixed Data and Task Parallelism on Heterogeneous Processor Grids (2002)

Unit Mixte, Olivier Beaumont, Arnaud Legrand, ...

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent...

Optimal Algorithms for Scheduling Divisible Workloads on Heterogeneous Systems (2002)

Unit Mixte, Arnaud Legrand, Olivier Beaumont, ...

In this paper, we discuss several algorithms for scheduling divisible loads on heterogeneous systems. Our main contributions are (i) new optimality results for single-round algorithms and (ii) the...

Bandwidth-Centric Allocation of Independent Tasks on Heterogeneous Platforms (2001)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Robert, Yves

In this paper, we consider the problem of allocating a large number of independent, equal-sized tasks to a heterogenerous "grid" computing platform. Such problems arise in collaborative computing...

The Master-Slave Paradigm with Heterogeneous Processors (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we revisit the master-slave tasking paradigm in the context of heterogeneous processors. We assume that communications take place in exclusive mode. We present a polynomial algorithm...

Data Allocation Strategies for Dense Linear Algebra on two-dimensional Grids with Heterogeneous Communication Links (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we study the implementation of dense linear algebra kernels, such as matrix multiplication on 2D grids with homogeneous processors when the communication links between the processors...

Bandwidth-Centric Allocation of Independent Tasks on Heterogeneous Platforms (2001)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Robert, Yves

In this paper, we consider the problem of allocating a large number of independent, equal-sized tasks to a heterogenerous "grid" computing platform. Such problems arise in collaborative computing...

The Master-Slave Paradigm with Heterogeneous Processors (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we revisit the master-slave tasking paradigm in the context of heterogeneous processors. We assume that communications take place in exclusive mode. We present a polynomial algorithm...

Data Allocation Strategies for Dense Linear Algebra on two-dimensional Grids with Heterogeneous Communication Links (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we study the implementation of dense linear algebra kernels, such as matrix multiplication on 2D grids with homogeneous processors when the communication links between the processors...

Bandwidth-Centric Allocation of Independent Tasks on Heterogeneous Platforms (2001)

Beaumont, Olivier, Carter, Larry, Ferrante, Jeanne, Legrand, Arnaud, Robert, Yves

In this paper, we consider the problem of allocating a large number of independent, equal-sized tasks to a heterogenerous "grid" computing platform. Such problems arise in collaborative computing...

The Master-Slave Paradigm with Heterogeneous Processors (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we revisit the master-slave tasking paradigm in the context of heterogeneous processors. We assume that communications take place in exclusive mode. We present a polynomial algorithm...

Data Allocation Strategies for Dense Linear Algebra on two-dimensional Grids with Heterogeneous Communication Links (2001)

Beaumont, Olivier, Legrand, Arnaud, Robert, Yves

In this paper, we study the implementation of dense linear algebra kernels, such as matrix multiplication on 2D grids with homogeneous processors when the communication links between the processors...

Alignment and distribution is NOT (always) NP-hard (2001)

Vincent Boudet Fabrice, Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Loop parallelization algorithms (2001)

Alain Darte, Yves Robert

Summary. This chapter is devoted to a comparative survey of loop parallelization algorithms. Various algorithms have been presented in the literature, such as those introduced by Allen and Kennedy,...

A Proposal for a Heterogeneous Cluster ScaLAPACK (Dense Linear Solvers) (2001)

Unite Mixte, Vincent Boudet, Fabrice Rastello, Yves Robert, Ecole Normale, ...

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). How is it possible to eÆciently implement...

Matrix Multiplication on Heterogeneous Platforms (2001)

Olivier Beaumont, Vincent Boudet, Fabrice Rastello, Yves Robert

this paper, we address the issue of implementing matrix multiplication on heterogeneous platforms. We target two different classes of heterogeneous computing resources: heterogeneous networks of...

A Proposal for a Heterogeneous Cluster ScaLAPACK (Dense Linear Solvers) (2001)

Unite Mixte, Vincent Boudet, Fabrice Rastello, Yves Robert

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). How is it possible to eÆciently implement...

The Master-Slave Paradigm with Heterogeneous Processors (2001)

École Normale, Supérieure Lyon, Olivier Beaumont, Arnaud Legrand, Arnaud Legrand, Yves ROBERT .

me amount of processing. . initiated). . exclusive mode. . Pool of independent tasks to be processed by the p slaves. . All tasks are of same-size, i.e. they represent the same amount of processing....

Bandwidth-Centric Allocation of Independent Tasks on Heterogeneous Platforms (2001)

Olivier Beaumont, Larry Carter, Jeanne Ferrante, Arnaud Legrand, Yves Robert

In this paper, we consider the problem of allocating a large number of independent, equalsized tasks to a heterogenerous "grid" computing platform. Such problems arise in collaborative...

Partitioning a square into rectangles: NP-completeness and approximation algorithms (2000)

Olivier Beaumont, Vincent Boudet, Fabrice Rastello, Yves Robert

In this paper, we deal with two geometric problems arising from heterogeneous parallel computing: how to partition the unit square into p rectangles of given area s 1; s 2; : : : ; s p (such that P p...

Partitioning a Square into Rectangles: NP-Completeness and Approximation Algorithms (2000)

Unite Mixte, Ecole Normale, Sup Lyon, Olivier Beaumont, Olivier Beaumont, ...

In this paper, we deal with two geometric problems arising from heterogeneous parallel computing: how to partition the unit square into p rectangles of given area s 1 ; s 2 ; : : : ; s p (such that P...

Partitioning a Square into Rectangles: NP-Completeness and Approximation Algorithms (2000)

Unite Mixte, Ecole Normale, Sup Lyon, Olivier Beaumont Vincent, Olivier Beaumont, ...

In this paper, we deal with two geometric problems arising from heterogeneous parallel computing: how to partition the unit square into p rectangles of given area s 1 ; s 2 ; : : : ; s p (such that P...

Heterogeneity Considered Harmful to Algorithm Designers (2000)

Unite Mixte, Ecole Normale, Sup Lyon, Olivier Beaumont, Olivier Beaumont, ...

In this paper, we deal with algorithmic issues on heterogeneous platforms. We show that static scheduling and load-balancing strategies are absolutely needed to achieve good performances, in contrast...

Load Balancing Strategies for Dense Linear Algebra Kernels on Heterogeneous Two-dimensional Grids (2000)

Olivier Beaumont, Vincent Boudet, Fabrice Rastello, Yves Robert

We study the implementation of dense linear algebra computations, such as matrix multiplication and linear system solvers, on two-dimensional (2D) grids of heterogeneous processors. For these...

Matrix-Matrix Multiplication on Heterogeneous Platforms (2000)

Unite Mixte, Ecole Normale, Sup Lyon, Olivier Beaumont, Olivier Beaumont, ...

In this paper, we address the issue of implementing matrix-matrix multiplication on heterogeneous platforms. We target two different classes of heterogeneous computing resources: heterogeneous...

Laboratoire de l'Informatique du Parall elisme (2000)

Unite Mixte, Ecole Normale Superieure, Sup Lyon, Olivier Beaumont, Olivier Beaumont, ...

In this paper, we deal with redistribution issues for dense linear algebra kernels on heterogeneous platforms. In this context, processors speeds may well vary during the execution of a large kernel,...

Static Tiling for Heterogeneous Computing Platforms (1999)

Pierre Boulet, Jack Dongarra, Yves Robert, Frédéric Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a sourceto -source program transformation. However, little work has been devoted to the mapping and scheduling of...

Algorithmic Issues for (Distributed) Heterogeneous Computing Platforms (Extended Abstract) (1999)

Unite Mixte, Ecole Normale, Sup Lyon, Vincent Boudet, Vincent Boudet, ...

Vincent Boudet, Fabrice Rastello and Yves Robert March 1999 Research Report N o 1999-19 Ecole Normale Sup erieure de Lyon 46 Allee d'Italie, 69364 Lyon Cedex 07, France Telephone :...

Data Allocation Strategies for Dense Linear Algebra Kernels on Heterogeneous Two-dimensional Grids. (1999)

Unite Mixte, Vincent Boudet Antoine, Ecole Normale, Sup Lyon, Antoine Petitet, ...

We study the implementation of dense linear algebra computations, such as matrix multiplication and linear system solvers, on two-dimensional (2D) grids of heterogeneous processors. For these...

Data Allocation Strategies for Dense Linear Algebra Kernels on Heterogeneous Two-dimensional Grids. (1999)

Unite Mixte, Vincent Boudet Antoine, Ecole Normale, Sup Lyon, Antoine Petitet, ...

We study the implementation of dense linear algebra computations, such as matrix multiplication and linear system solvers, on two-dimensional (2D) grids of heterogeneous processors. For these...

Data Allocation Strategies for Dense Linear Algebra Kernels on Heterogeneous Two-dimensional Grids. (1999)

Unite Mixte, Vincent Boudet, Ecole Normale, Sup Lyon, Antoine Petitet, ...

We study the implementation of dense linear algebra computations, such as matrix multiplication and linear system solvers, on two-dimensional (2D) grids of heterogeneous processors. For these...

Data Allocation Strategies for Dense Linear Algebra Kernels on Heterogeneous Two-dimensional Grids (1999)

Vincent Boudet, Antoine Petitet, Fabrice Rastello, Yves Robert

We study the implementation of dense linear algebra computations, such as matrix multiplication and linear system solvers, on two-dimensional (2D) grids of heterogeneous processors. For these...

More on Scheduling Block-Cyclic Array Redistribution (1998)

Desprez, Frédéric, Domas, Stéphane, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper, we have reported how to...

More on Scheduling Block-Cyclic Array Redistribution (1998)

Desprez, Frédéric, Domas, Stéphane, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper, we have reported how to...

More on Scheduling Block-Cyclic Array Redistribution (1998)

Desprez, Frédéric, Domas, Stéphane, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper, we have reported how to...

Tiling for heterogeneous computing platforms (1998)

Pierre Boulet, Jack Dongarra, Yves Robert, Frederic Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a source-to-source program transformation. However, little work has been devoted to the mapping and scheduling of...

Alignment and Distribution Is Not (Always) (1998)

Vincent Boudet, Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to simultaneously implement array alignment and data computation distribution is introduced and evaluated. We

Scheduling block-cyclic array redistribution (1998)

Frederic Desprez, Jack Dongarra, Antoine Petitet, Cyril R, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on e...

SPI Loop Partitioning versus Tiling for Cache-based Multiprocessors (1998)

Fabrice Rastello, Yves Robert, Fabrice Rastello, Yves Robert, Fabrice Rastello, Yves Robert

In this paper, an e cient algorithm to implement loop partitioning is introduced and evaluated. We improve recent results of Agarwal, Kranz and Natarajan [1] inseveral directions. We give a more...

Alignment and distribution is not (always) np--hard (1998)

Vincent Boudet, Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Circuit Retiming Applied to Decomposed Software Pipelining (1998)

Pierre-Yves Calland, Alain Darte, Yves Robert

This paper elaborates on a new view on software pipelining, called decomposed software pipelining, and introduced by Gasperoni and Schwiegelshohn, and by Wang, Eisenbeis, Jourdan, and Su. The...

Alignment and distribution is NOT (always) NP-hard (1998)

Vincent Boudet Fabrice, Ecole Normale, Sup Lyon, Fabrice Rastello, Fabrice Rastello, Yves Robert, ...

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Alignment and distribution is NOT (always) NP-hard (1998)

Vincent Boudet Fabrice, Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [12, 14], and...

More on Scheduling Block-Cyclic Array Redistribution (1998)

Frédéric Desprez, Unite Mixte, Ecole Normale, Sup Lyon, ...

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. In a previous paper, we have reported how to...

Technology Transfer within the ProHPC TTN at ENS Lyon (1998)

Prohpc Ttn, Ens Lyon, Frédéric Desprez, Christophe Barberet, Brunie Fr'ed'eric Desprez, Stéphane Ubéda, ...

This article is devoted to the description of our activities related to transferring the HPCN technology to SMEs. This work is performed in the framework of the French TTN ProHPC which we briefly...

Laboratoire de l'Informatique du Parallelisme (1998)

Ecole Normale Sup'erieure, Vincent Boudet, Ecole Normale, Sup Lyon, Fabrice Rastello, Fabrice Rastello, ...

this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated

Algorithmic Issues on Heterogeneous Computing Platforms (1998)

Pierre Boulet, Jack Dongarra, Fabrice Rastello, Yves Robert, Frédéric Vivien

This paper discusses some algorithmic issues when computing with a heterogeneous network of workstations (the typical poor man's parallel computer). Dealing with processors of dierent speeds...

Tiling for Heterogeneous Computing Platforms (1998)

Ecole Normale, Suprieure Lyon, Pierre Boulet, Frédéric Vivien, Pierre Boulet, Jack Dongarra, ...

In the framework of fully permutable loops, tiling has been extensively studied as a source-to-source program transformation. However, little work has been devoted to the mapping and scheduling of...

Loop Partitioning versus Tiling for Cache-based Multiprocessors (1998)

Ecole Normale, Sup Lyon, Fabrice Rastello, Fabrice Rastello, Yves Robert, Yves Robert

In this paper, an efficient algorithm to implement loop partitioning is introduced and evaluated. We improve recent results of Agarwal, Kranz and Natarajan [1] in several directions. We give a more...

Loop Partitioning for Cache-based Multiprocessors (1998)

Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to implement loop partitioning is introduced and evaluated. We improve recent results of Agarwal, Kranz and Natarajan [1] in several directions. We derive a new...

Alignment and distribution is NOT (always) NP-hard (1998)

Vincent Boudet, Fabrice Rastello, Yves Robert

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Loop Partitioning versus Tiling for Cache-based Multiprocessors (1998)

Fabrice Rastello, Fabrice Rastello, Yves Robert, Yves Robert

In this paper, an efficient algorithm to implement loop partitioning is introduced and evaluated. We improve recent results of Agarwal, Kranz and Natarajan [1] in several directions. We give a more...

Alignment and distribution is NOT (always) NP-hard (1998)

Vincent Boudet, Ecole Normale, Sup Lyon, Fabrice Rastello, Fabrice Rastello, Yves Robert, ...

In this paper, an efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Circuit Retiming Applied to Decomposed Software Pipelining (1998)

Pierre-Yves Calland, Alain Darte, Yves Robert

This paper elaborates on a new view on software pipelining, called decomposed software pipelining, and introduced by Gasperoni and Schwiegelshohn, and by Wang, Eisenbeis, Jourdan, and Su. The...

Scheduling block-cyclic array redistribution (1998)

Frederic Desprez, Jack Dongarra, Antoine Petitet, Cyril R, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on e...

Scheduling block-cyclic array redistribution (1998)

Frederic Desprez, Jack Dongarra, Antoine Petitet, Cyril R, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on e...

Alignment and distribution is not (always) np--hard (1998)

Vincent Boudet, Fabrice Rastello, Fabrice Rastello, Yves Robert, Yves Robert

In this paper, an e cient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen [13, 14], and...

Determining the Idle Time of a Tiling: New Results (1997)

Desprez, Frédéric, Dongarra, Jack, Rastello, Fabrice, Robert, Yves

In the framework of fully permutable loops, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Högsted, Carter, and...

Tiling with Limited Resources (1997)

Calland, Pierre-Yves, Dongarra, Jack, Robert, Yves

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformati- on. Little work has been devoted to the mapping and...

Scheduling Block-Cyclic Array Redistribution (1997)

Desprez, Frédéric, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of arrays that are distributed in a block-cyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Determining the Idle Time of a Tiling: New Results (1997)

Desprez, Frédéric, Dongarra, Jack, Rastello, Fabrice, Robert, Yves

In the framework of fully permutable loops, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Högsted, Carter, and...

Tiling with Limited Resources (1997)

Calland, Pierre-Yves, Dongarra, Jack, Robert, Yves

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformati- on. Little work has been devoted to the mapping and...

Scheduling Block-Cyclic Array Redistribution (1997)

Desprez, Frédéric, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of arrays that are distributed in a block-cyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Determining the Idle Time of a Tiling: New Results (1997)

Desprez, Frédéric, Dongarra, Jack, Rastello, Fabrice, Robert, Yves

In the framework of fully permutable loops, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Högsted, Carter, and...

Tiling with Limited Resources (1997)

Calland, Pierre-Yves, Dongarra, Jack, Robert, Yves

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformati- on. Little work has been devoted to the mapping and...

Scheduling Block-Cyclic Array Redistribution (1997)

Desprez, Frédéric, Dongarra, Jack, Petitet, Antoine, Randriamaro, Cyril, Robert, Yves

This article is devoted to the run-time redistribution of arrays that are distributed in a block-cyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Determining the idle time of a tiling: new results (1997)

Frederic Desprez, Frederic Desprez, Jack Dongarra, Jack Dongarra, Fabrice Rastello, Fabrice Rastello, ...

In the framework of fully permutable loops, tiling has been studied extensively as a sourceto-source program transformation. We build upon recent results by Hogsted, Carter, and Ferrante [12], who...

Tiling for heterogeneous computing platforms (1997)

Pierre Boulet, Jack Dongarra, Yves Robert, Frédéric Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a sourceto-source program transformation. However, little work has been devoted to the mapping and scheduling of the...

Tiling for Heterogeneous Computing Platforms (1997)

Pierre Boulet, Jack Dongarra, Yves Robert, Fr'ed'eric Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a sourceto -source program transformation. However, little work has been devoted to the mapping and scheduling of...

Plugging Anti and Output Dependence Removal Techniques Into Loop Parallelization Algorithms (1997)

Frdric Vivien, Pierre-yves Call, Alain Darte, Alain Darte, Yves Robert, ...

: In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such "false" dependences. These false...

Tiling on Systems with Communication/Computation Overlap (1997)

Pierre-yves Calland, Jack Dongarra, Yves Robert

In the framework of fully permutable loops, tiling is a compiler technique (also known as \loop blocking") that has been extensively studied as a source-to-source program transformation. Little...

Tiling With Limited Resources (1997)

Pierre-yves Call, Jack Dongarra, Jack Dongarra, Yves Robert, Yves Robert, ...

: In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Determining the Idle Time of a Tiling: New Results (1997)

Frederic Desprez Jack, Jack Dongarra, Jack Dongarra, Fabrice Rastello, Fabrice Rastello, Yves Robert, ...

: In the framework of fully permutable loops, tiling has been studied extensively as a source-tosource program transformation. We build upon recent results by Hogsted, Carter, and Ferrante [12], who...

Scheduling Block-Cyclic Array Redistribution (1997)

Frédéric Desprez, Th Eme, Jack Dongarra, Cyril Randriamaro, Jack Dongarra, Antoine Petitet, ...

: This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Plugging Anti and Output Dependence Removal Techniques Into Loop Parallelization Algorithms (1997)

Pierre-Yves Calland, Alain Darte, Yves Robert, Frédéric Vivien

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such "false" dependences. These false dependences...

Determining the Idle Time of a Tiling: New Results (1997)

Frédéric Desprez, Ecole Normale, Suprieure Lyon, Jack Dongarra, Jack Dongarra, Fabrice Rastello, ...

In the framework of fully permutable loops, tiling has been studied extensively as a sourceto -source program transformation. We build upon recent results by Hogsted, Carter, and Ferrante [12], who...

Determining the Idle Time of a Tiling: New Results (1997)

Frederic Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of perfect loop nests with uniform dependencies, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Hogsted, Carter,...

Tiling with Limited Resources (1997)

Pierre-Yves Calland, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Determining the Idle Time of a Tiling: New Results (1997)

Fr'ed'eric Desprez Jack, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of fully permutable loops, tiling has been studied extensively as a sourceto -source program transformation. We build upon recent results by Hogsted, Carter, and Ferrante [12], who...

Scheduling Block-Cyclic Array Redistribution (1997)

Frédéric Desprez, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of one-dimensional arrays that are distributed in a block-cyclic fashion over a processor grid. While previous studies have concentrated on...

Scheduling Block-Cyclic Array Redistribution (1997)

Fr'ed'eric Desprez Jack, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Scheduling Block-Cyclic Array Redistribution (1997)

Fr'ed'eric Desprez Jack, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a block-cyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Tiling with Limited Resources (1997)

Pierre-yves Calland, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Determining the Idle Time of a Tiling: New Results (1997)

Frédéric Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of perfect loop nests with uniform dependencies, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Hogsted, Carter,...

Scheduling Block-Cyclic Array Redistribution (1997)

Fr'ed'eric Desprez, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Tiling With Limited Resources (1997)

Pierre-yves Calland, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms (1997)

Pierre-yves Calland, Alain Darte, Yves Robert, Frederic Vivien, Eric Vivien

. In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such "false" dependences. These false...

Scheduling Block-Cyclic Array Redistribution (1997)

Frédéric Desprez, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, Yves Robert

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms (1997)

Pierre-yves Calland, Alain Darte, Yves Robert, Frederic Vivien

this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such "false" dependences. These false dependences are...

Tiling with Limited Resources (1997)

Pierre-yves Calland, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Tiling for Heterogeneous Computing Platforms (1997)

Pierre Boulet, Jack Dongarra, Yves Robert, Frédéric Vivien, Fr'ed'eric Vivien

In the framework of fully permutable loops, tiling has been extensively studied as a sourceto -source program transformation. However, little work has been devoted to the mapping and scheduling of...

Tiling With Limited Resources (1997)

Pierre-yves Calland, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Determining the Idle Time of a Tiling: New Results (1997)

Frédéric Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of fully permutable loops, tiling has been studied extensively as a source-to-source program transformation. We build upon recent results by Hogsted, Carter, and Ferrante [12], who...

Scheduling Block-Cyclic Array Redistribution (1997)

Fr'ed'eric Desprez, Frederic Desprez, Cyril R, Jack Dongarra, Jack Dongarra, Antoine Petitet, ...

This article is devoted to the run-time redistribution of arrays that are distributed in a blockcyclic fashion over a multidimensional processor grid. While previous studies have concentrated on...

Determining the Idle Time of a Tiling: New Results (1997)

Frédéric Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. We build upon recent results by Hogsted, Carter and...

Determining the Idle Time of a Tiling: New Results (1997)

Fr'ed'eric Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. We build upon recent results by Hogsted, Carter and...

Tiling with limited resources (1997)

Pierre-yves Call, Jack Dongarra, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and...

Determining the idle time of a tiling: new results (1997)

Frederic Desprez, Jack Dongarra, Fabrice Rastello, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling has been extensively studied as a source-to-source program transformation. We build upon recent results byHogsted, Carter and...

Tiling on Systems with Communication/Computation Overlap (1997)

Pierre-Yves Calland, Jack Dongarra, Yves Robert

In the framework of fully permutable loops, tiling is a compiler technique (also known as "loop blocking") that has been extensively studied as a source-to-source program transformation....

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such «false» dependences. These false dependences are...

On the Removal of Anti and Output Dependences (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we build upon results of Padua and Wolfe~\cite{PaduaWo86}, who introduce two graph transformations to eliminate anti and output dependences. We first give a unified framework for such...

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such «false» dependences. These false dependences are...

On the Removal of Anti and Output Dependences (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we build upon results of Padua and Wolfe~\cite{PaduaWo86}, who introduce two graph transformations to eliminate anti and output dependences. We first give a unified framework for such...

Plugging Anti and Output Dependence Removal Techniques into Loop Parallelization Algorithms (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such «false» dependences. These false dependences are...

On the Removal of Anti and Output Dependences (1996)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves, Vivien, Frédéric

In this paper we build upon results of Padua and Wolfe~\cite{PaduaWo86}, who introduce two graph transformations to eliminate anti and output dependences. We first give a unified framework for such...

On the removal of anti and output dependences (1996)

Ecole Normale, Supérieure Lyon, Alain Darte, Yves Robert, Frederic Vivien, Pierre-yves Calland, ...

Unité de recherche associée au CNRS n°1398 On the removal of anti and output dependences

Resource-constrained scheduling of partitioned algorithms on processor arrays (1996)

Michele Dion, Michele Dion, Michele Dion, Tanguy Risset, Tanguy Risset, Tanguy Risset, ...

We deal with the problem of partitioning and mapping uniform loop nests onto physical processor arrays. Resource constraints are taken into account: not only we assume a limited number of available...

A characterization of one-to-one modular mappings (1996)

Alain Darte, Michele Dion, Yves Robert, Alain Darte, Michele Dion, Yves Robert, ...

In this paper, we deal with modular mappings as introduced by LeeandFortes [14, 13, 12], and we build upon their results. Our main contribution is a characterization of one-to-one modular mappings...

On the removal of anti and output dependences (1996)

Pierre-yves Calland, Alain Darte, Yves Robert

In this paper we build upon results of Padua and Wolfe [9], who introduced two graph transformations to break dependence paths including anti and output dependences. We first formalize these two...

(Pen)-ultimate tiling? (1996)

Pierre Boulet, Alain Darte, Tanguy Risset, Yves Robert

In the framework of perfect loop nests with uniform dependences, tiling is a technique used to group elemental computation points so as to increase computation granularity and to reduce the overhead...

On the Removal of Anti and Output Dependences (1996)

Ecole Normale, Suprieure Lyon, Pierre-yves Calland, Frédéric Vivien, Pierre-yves Calland, Alain Darte, ...

In this paper we build upon results of Padua and Wolfe [8], who introduce two graph transformations to eliminate anti and output dependences. We first give a unified framework for such...

Plugging Anti and Output Dependence Removal Techniques Into Loop Parallelization Algorithms (1996)

Ecole Normale, Suprieure Lyon, Pierre-yves Calland, Pierre-yves Calland, Alain Darte, Alain Darte, ...

In this paper we shortly survey some loop transformation techniques which break anti or output dependences, or artificial cycles involving such "false" dependences. These false dependences...

On the Removal of Anti and Output Dependences (1996)

Alain Darte, Yves Robert

In this paper we build upon results of Padua and Wolfe [9], who introduced two graph transformations to break dependence paths including anti and output dependences. We first formalize these two...

On the Removal of Anti and Output Dependences (1996)

Pierre-yves Calland, Alain Darte, Yves Robert, Frédéric Vivien

In this paper we build upon results of Padua and Wolfe [9], who introduced two graph transformations to break dependence paths including anti and output dependences. We rst formalize these two...

Resource-constrained Scheduling of Partitioned Algorithms on Processor Arrays (1996)

Michèle Dion, Tanguy Risset, Yves Robert

We deal with the problem of partitioning and mapping uniform loop nests onto physical processor arrays. Resource constraints are taken into account: not only we assume a limited number of available...

Mapping Affine Loop Nests (1996)

Michèle Dion, Yves Robert

This paper deals with the problem of aligning data and computations when mapping affine loop nests onto Distributed Memory Parallel Computers (DMPCs). We formulate the problem by introducing a new...

Compiling Affine Nested Loops: How to Optimize the Residual Communications After the Alignment Phase? (1996)

Michèle Dion, Cyril Randriamaro, Cyril R, Yves Robert

Minimizing communication overhead when mapping affine loop nests onto distributed memory parallel computers (DMPCs) is a key problem with regard to performance, and many authors have dealt with it....

How to optimize residual communications (1996)

Michèle Dion, Cyril R, Yves Robert

Minimizing communications when mapping affine loop nests onto distributed memory parallel computers has already drawn a lot of attention. This paper focuses on the next step: as it is generally...

A New Guaranteed Heuristic for the Software Pipelining Problem (1995)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves

We present yet another heuristic for the software pipelining problem. We believe this heuristic to be of interest because it brings a new insight to the software pipelining problem by establishing...

A New Guaranteed Heuristic for the Software Pipelining Problem (1995)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves

We present yet another heuristic for the software pipelining problem. We believe this heuristic to be of interest because it brings a new insight to the software pipelining problem by establishing...

A New Guaranteed Heuristic for the Software Pipelining Problem (1995)

Calland, Pierre-Yves, Darte, Alain, Robert, Yves

We present yet another heuristic for the software pipelining problem. We believe this heuristic to be of interest because it brings a new insight to the software pipelining problem by establishing...

A new guaranteed heuristic for the software pipelining problem (1995)

Ecole Normale, Supérieure Lyon, Alain Darte, Yves Robert, Pierre-yves Calland, Alain Darte, ...

Unité de recherche associée au CNRS n°1398 A new guaranteed heuristic for the software pipelining problem

How to optimize residual communications? (1995)

Ecole Normale, Supérieure Lyon, Michele Dion, Cyril Randriamaro, Yves Robert, Michele Dion, ...

Unité de recherche associée au CNRS n°1398 How to optimize residual communications?

A New Guaranteed Heuristic for the Software Pipelining Problem (1995)

Ecole Normale, Suprieure Lyon, Pierre-yves Calland, Pierre-yves Calland, Alain Darte, Alain Darte, ...

We present yet another heuristic for the software pipelining problem. We believe this heuristic to be of interest because it brings a new insight to the software pipelining problem by establishing...

A New Guaranteed Heuristic for the Software Pipelining Problem (1995)

Pierre-yves Call, Alain Darte, Alain Darte, Yves Robert, Yves Robert

: We present yet another heuristic for the software pipelining problem. We believe this heuristic to be of interest because it brings a new insight to the software pipelining problem by establishing...

Affine-by-Statement Scheduling of Uniform and Affine Loop Nests over Parametric Domains (1995)

Alain Darte, Yves Robert

This paper deals with parallel scheduling techniques for uniform and affine loop nests. We deal with affine-by-statement scheduling, a powerful extension of Lamport's hyperplane method where...

Evaluating Array Expressions on Massively Parallel Machines with Communication/Computation Overlap (1995)

Vincent Bouchitté, Pierre Boulet, Alain Darte, Yves Robert

This paper deals with the problem of evaluating HPF style array expressions on massively parallel distributed-memory computers (DMPCs). This problem has been addressed by Chatterjee et al. under the...

Laboratoire de l’Informatique du Parallélisme Ecole Normale Supérieure de Lyon (1995)

Pierre-yves Calland, Anne Mignotte, Olivier Peyran, Yves Robert, Frederic Vivien, Retiming Dags, ...

The increasing complexity of digital circuitry makes global design optimization no longer possible: a designer will only consider the critical parts of his circuit. This paper discusses timing...

On The Alignment Problem (1994)

Alain Darte, Yves Robert

This paper deals with the problem of aligning data and computations when mapping uniform or affine loop nests onto SPMD distributed memory parallel computers. For affine loop nests we formulate the...

Constructive Methods for Scheduling Uniform Loop Nests (1994)

Alain Darte, Yves Robert

This paper surveys scheduling techniques for loop nests with uniform dependences. First we introduce the hyperplane method and related variants. Then we extend it by using a different affine...

Affine-by-Statement Scheduling of Uniform Loop Nests over Parametric Domains (1993)

Alain Darte, Yves Robert

this report we deal with affine-by-statement scheduling, a high-level technique for the parallelization of loop nests with uniform dependences. Affineby -statement scheduling can be viewed as a...

Scheduling Uniform Loop Nests (1992)

Alain Darte, Yves Robert

This paper surveys scheduling techniques for uniform loop nests. First we introduce the hyperplane method and related variants. Then we extend it by using a different affine scheduling for each...

Revisiting Cycle Shrinking (1992)

Yves Robert, Siang W. Song

Several loop transformations techniques have been designed to extract parallelism from nested loop structures. We first review two important approaches, known as Generalized Cycle Shrinking presented...

Linear Scheduling Is Nearly Optimal (1991)

Alain Darte, Leonid Khachiyan, Yves Robert

This paper deals with the problem of finding optimal schedulings for uniform dependence algorithms. Given a convex domain, let T f be the total time needed to execute all computations using the free...

Linear Scheduling is Nearly Optimal (1991)

Alain Darte, Leonid Khachiyan, Yves Robert, Lyon Cedex

This paper deals with the problem of finding optimal schedulings for uniform dependence algorithms. Given a convex domain, let T f be the total time needed to execute all computations using the free...

Systolic convolution of arithmetic functions (1989)

Quinton, Patrice, Robert, Yves

Disponible dans les fichiers attachés à ce document

Systolic convolution of arithmetic functions (1989)

Quinton, Patrice, Robert, Yves

Disponible dans les fichiers attachés à ce document

Algorithmique parallèle : réseaux d'automates, architectures systoliques, machines SIMD & MIMD / (1986)

Robert, Yves.

Thèse (Docteur es sciences)--Institut national polytechnique de Grenoble, Université scientifique et médicale de Grenoble, 1986.

Quelques algorithmes systoliques pour le calcul scientifique / (1982)

Robert, Yves.

Thèse (Ph. D.)--Université scientifique et médicale de Grenoble, 1982.