Despite recent advances in compiler technology, programmers are still confronted with machinedependent optimization issues such as code and data placement in order to fully exploit a target parallel machine's performance potential. Reconciling the generally conflicting goals of portability and performance requires completely automatic, retargetable program compilation. The efficiency of the cost estimation techniques needed to steer the compile-time optimization process, however, critically depends on the choice of parallel programming model. We present a new coordination model, called SPC, that imposes specific restrictions in the synchronization structures that can be programmed. Imposing these restrictions enables the efficient comp...
Abstract. Helping programmers write parallel software is an urgent problem given the popularity of m...
International audienceThis paper presents a technique for representing the high level semantics of p...
Most of the current compiler projects for distributed memory architectures leave the critical and ti...
. A new model of parallel computation is presented that provides appropriate cost models for automat...
. We present compiler optimization techniques for explicitly parallel programs that communicate thro...
INTRODUCTION The SPMD (Single-Program Multiple-Data Stream) model has been widely adopted as the ba...
The goal of this dissertation is to give programmers the ability to achieve high performance by focu...
Efficiently using multicore architectures demands an increasing degree of fluency in parallel progra...
An ideal language for parallel programming will have to satisfy simultaneously many conflicting requ...
In this paper, we focus on the need for two approaches to optimize producer and consumer synchroniza...
This work was also published as a Rice University thesis/dissertation: http://hdl.handle.net/1911/16...
In this paper we analyze the effect of compiler optimizations on fine grain parallelism in scalar pr...
Abstract. Application performance is heavily dependent on the com-piler optimizations. Modern compil...
Generating optimal code is a challenging problem. Traditional compilers break down the problem compl...
Power consumption and fabrication limitations are increasingly playing significant roles in the desi...
Abstract. Helping programmers write parallel software is an urgent problem given the popularity of m...
International audienceThis paper presents a technique for representing the high level semantics of p...
Most of the current compiler projects for distributed memory architectures leave the critical and ti...
. A new model of parallel computation is presented that provides appropriate cost models for automat...
. We present compiler optimization techniques for explicitly parallel programs that communicate thro...
INTRODUCTION The SPMD (Single-Program Multiple-Data Stream) model has been widely adopted as the ba...
The goal of this dissertation is to give programmers the ability to achieve high performance by focu...
Efficiently using multicore architectures demands an increasing degree of fluency in parallel progra...
An ideal language for parallel programming will have to satisfy simultaneously many conflicting requ...
In this paper, we focus on the need for two approaches to optimize producer and consumer synchroniza...
This work was also published as a Rice University thesis/dissertation: http://hdl.handle.net/1911/16...
In this paper we analyze the effect of compiler optimizations on fine grain parallelism in scalar pr...
Abstract. Application performance is heavily dependent on the com-piler optimizations. Modern compil...
Generating optimal code is a challenging problem. Traditional compilers break down the problem compl...
Power consumption and fabrication limitations are increasingly playing significant roles in the desi...
Abstract. Helping programmers write parallel software is an urgent problem given the popularity of m...
International audienceThis paper presents a technique for representing the high level semantics of p...
Most of the current compiler projects for distributed memory architectures leave the critical and ti...