Esodyp+: Prefetching in the Jackal Software DSM

Klemm, Michael; Beyler, Jean Christophe; Lampert, Ronny T.; Philippsen, Michæl; Clauss, Philippe

doi:10.1007/978-3-540-74466-5_60

Cited by 5 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the accuracy of the Stride predictor is below a certain threshold, the second predictor is used. This second predictor is the Markov model predictor Esodyp+ [7], which uses past events to predict future events. It relies on the observation that events of the past are likely to repeat.…”

Section: Profiling State Machinementioning

confidence: 99%

See 1 more Smart Citation

A meta‐predictor framework for prefetching in object‐based DSMs

Beyler

Klemm

Clauss³

et al. 2009

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SUMMARYDynamic optimizers modify the binary code of programs at runtime by profiling and optimizing certain aspects of the execution. We present a completely software-based framework that dynamically optimizes programs for object-based distributed shared memory (DSM) systems on clusters. In DSM systems, reducing the number of messages between cluster nodes is crucial. Prefetching transfers data in advance from the storage node to the local node so that communication is minimized. Our framework uses a profiler and a dynamic binary rewriter that monitor the access behavior of the application and place prefetches where they are beneficial to speed up the application. In addition, we use two distinct predictors to handle different types of access patterns. A meta-predictor analyzes the memory access behavior and dynamically enables one of the predictors. Our system also adapts the number of prefetches per request to best fit the application's behavior. The evaluation shows that the performance of our system is better than the manual prefetching. The number of messages sent decreases by up to 90%. Performance gains of up to 80% can be observed on benchmarks.

show abstract

Section: Profiling State Machinementioning

confidence: 99%

“…The Esodyp+ predictor [7] extends the Entirely Software DYnamic data Prefetcher (Esodyp), a predictor that modelizes address sequences with a variation of the classic Markov model [12]. This section briefly presents the model, how it reacts to the addresses passed by the DSM framework, and how it helps prefetch objects.…”

Section: Esodyp+mentioning

confidence: 99%

A meta‐predictor framework for prefetching in object‐based DSMs

Beyler

Klemm

Clauss³

et al. 2009

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the Model Creation state and the Prefetching state we use the Markov model predictor Esodyp+ [10], which uses past events to predict future events. It relies on the observation that events of the past are likely to repeat.…”

Section: Profiling State Machinementioning

confidence: 99%

“…However, if profitable, the code rewriter replaces the monitoring calls with prefetcher calls. The Esodyp+ prefetcher [10] is used as an example for our generic optimizer. It predicts memory access patterns and prefetches future memory accesses.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Prefetching with Binary Code Rewriting in Object-Based DSMs

Beyler¹,

Klemm

Philippsen

et al. 2008

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Dynamic optimizers modify the binary code of programs at runtime by profiling and optimizing certain aspects of the execution. We present a completely software-based framework that dynamically optimizes programs for object-based Distributed Shared Memory (DSM) systems. In DSM systems, reducing the number of messages between nodes is crucial. Prefetching transfers data in advance from the storage node to the local node so that communication is minimized. Our framework uses a profiler and a dynamic binary rewriter that monitors the access behavior of the application and places prefetches where they are beneficial to speed up the application. In addition, we adapt the number of prefetches per request to best fit the application's behavior. Evaluation shows that the performance of our system is better than manual prefetching. The number of messages sent decreases by up to 89%. Performance gains of up to 73% can be observed on the benchmarks.

show abstract