Applicability of the ECM Performance Model to Explicit ODE Methods on Current Multi-core Processors

Seiferth, Johannes; Alappat, Christie L.; Korch, Matthias; Rauber, Thomas

doi:10.1007/978-3-319-92040-5_9

Cited by 9 publications

(12 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we propose Offsite, an offline AT approach that automatically identifies the most efficient implementation variant(s) during installation time based on performance predictions. These predictions stem from an analytic performance prediction methodology for explicit ODE methods proposed by [19] that uses a combined white and black box model approach based on the ECM model. The main contributions of this paper are:…”

Section: Main Contributionsmentioning

confidence: 99%

Offsite Autotuning Approach

Seiferth

Korch

Rauber

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Autotuning (AT) is a promising concept to minimize the often tedious manual effort of optimizing scientific applications for a specific target platform. Ideally, an AT approach can reliably identify the most efficient implementation variant(s) for a new platform or new characteristics of the input by applying suitable program transformations and analytic models. In this work, we introduce Offsite, an offline AT approach that automates this selection process at installation time by rating implementation variants based on an analytic performance model without requiring time-consuming runtime tests. From abstract multilevel description languages, Offsite automatically derives optimized, platform-specific and problem-specific code of possible variants and applies the performance model to these variants.We apply Offsite to parallel numerical methods for ordinary differential equations (ODEs). In particular, we investigate tuning a specific class of explicit ODE solvers, PIRK methods, for four different initial value problems (IVPs) on three different shared-memory systems. Our experiments demonstrate that Offsite can reliably identify the set of most efficient implementation variants for different given test configurations (ODE solver, IVP, platform) and effectively handle important AT scenarios.

show abstract

Section: Main Contributionsmentioning

confidence: 99%

Offsite Autotuning Approach

Seiferth

Korch

Rauber

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…With estimates for individual kernels in place we can now present multicore-scaling data for the full PCG algorithm. Composing the model from single-loop predictions is simple due to the time-based formulation of the ECM model [21]. In the case of PCG we have three invocations of daxpby, two of dot, one gs forward-and backward-sweep each, as well as one of stencil.…”

Section: Compositionmentioning

confidence: 99%

“…Both have been subject to intense study, refinements, and validation, and their areas of applicability are well understood. However, while there is ample data available for Roofline on a wide variety of architectures [15,18], one drawback of previous applications of the ECM model [2,4,12,21,22,24,26] is that they were mostly restricted to Intel processors. We provide the first thorough cross-architecture study of the model.…”

Section: Related Workmentioning

confidence: 99%

Bridging the Architecture Gap: Abstracting Performance-Relevant Properties of Modern Server Processors

Hofmann

Alappat

Hager

et al. 2020

JSFI

View full text Add to dashboard Cite

We propose several improvements to the execution-cache-memory (ECM) model, an analytic performance model for predicting single-and multicore runtime of steady-state loops on server processors. The model is made more general by strictly differentiating between application and machine models: an application model comprises the loop code, problem sizes, and other runtime parameters, while a machine model is an abstraction of all performance-relevant properties of a processor. Moreover, new first principles underlying the model's estimates are derived from common microarchitectural features implemented by today's server processors to make the model more architecture independent, thereby extending its applicability beyond Intel processors. We introduce a generic method for determining machine models, and present results for relevant server-processor architectures by Intel, AMD, IBM, and Marvell/Cavium. Considering this wide range of architectures, the set of features required for adequate performance modeling is surprisingly small. To validate our approach, we compare performance predictions to empirical data for an OpenMP-parallel preconditioned CG algorithm, which includes compute-and memory-bound kernels. Both single-and multicore analysis shows that the model exhibits average and maximum relative errors of 5 % and 10 %. Deviations from the model and insights gained are discussed in detail.

show abstract

“…In the online phase we can measure the runtime of the fastest specialized implementation variant and avoid the execution of the general implementation variants if their synchronization overhead is expected to be higher than the runtime of the best specialized implementation. Another idea is to use performance models in the offline phase, such as the ECM model, in order to predict the execution time of variants or at least to forecast their performance ranking [32]. Figure 8 demonstrates that for variants with loop tiling an appropriate selection of the tile size is important to achieve a good performance.…”

Section: Experimental Setup and Evaluationmentioning

confidence: 99%

“…Furthermore, the offline phase can be used to estimate synchronization and communication overheads for different configurations and input scenarios with the help of micro-benchmarks. This information can be used together with a performance model to predict a performance ranking of different implementation variants [32] and different parameter values. All evaluated configurations can be arranged in a decision tree according to their ranking, which is then forwarded to the online tuning phase.…”

Section: Summary Of the Observations For Ode Solversmentioning

confidence: 99%

A performance- and energy-oriented extended tuning process for time-step-based scientific applications

et al. 2020

Self Cite

View full text Add to dashboard Cite

Scientific application codes are often long-running time-and energy-consuming parallel codes, and the tuning of these methods towards the characteristics of a specific hardware is essential for a good performance. However, since scientific software is often developed over many years, the application software usually survives several hardware generations, which might make a re-tuning of the existing codes necessary. To simplify the tuning process, it would be beneficial to have software with inherent tuning possibilities. In this article, we explore the possibilities of tuning methods for time-step-based applications. Two different time-step-based application classes are considered, which are solution methods for ordinary differential equations and particle simulation methods. The investigation comprises a broad range of tuning possibilities, starting from the choice of algorithms, the parallel programming model, static implementation variants, input characteristics as well as hardware parameters for parallel execution. An experimental investigation shows the different characteristics of the application classes on different multicore systems. The results show that a combination of offline and online tuning leads to good tuning results. However, due to the different input characteristics of the two application classes, regular versus irregular, different tuning aspects are most essential.

show abstract

Applicability of the ECM Performance Model to Explicit ODE Methods on Current Multi-core Processors

Cited by 9 publications

References 21 publications

Offsite Autotuning Approach

Offsite Autotuning Approach

Bridging the Architecture Gap: Abstracting Performance-Relevant Properties of Modern Server Processors

A performance- and energy-oriented extended tuning process for time-step-based scientific applications

Contact Info

Product

Resources

About