Modeling Power and Energy Usage of HPC Kernels

Tiwari, Ananta; Laurenzano, Michael A.; Carrington, Laura; Snavely, Allan

doi:10.1109/ipdpsw.2012.121

Cited by 46 publications

(18 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In paper [6] authors used neural networks to train models that predicted power and energy consumption when running high performance computing codes. It has been shown that after training, using various versions of codes, it is possible to predict power consumption and energy usage of CPUs and DIMMs with less that 5.5% error for LU factorization, Jacobi and matrix multiplication.…”

Section: Related Workmentioning

confidence: 99%

Modeling energy consumption of parallel applications

Czarnul

Kuchta

Rościszewski

et al. 2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-The paper presents modeling and simulation of energy consumption of two types of parallel applications: geometric Single Program Multiple Data (SPMD) and divide-and-conquer (DAC). Simulation is performed in a new MERPSYS (Modeling Efficiency, Reliability and Power consumption of multilevel parallel HPC SYStems using CPUs and GPUs) environment. Model of an application uses the Java language with extensions representing message exchange between processes working in parallel. Simulation is performed by running threads representing distinct process codes of an application, with consideration of process counts. Instead of running time consuming calculations, their times are simulated using functions representing computational time dependent on input data sizes. The simulator considers performance and power consumption values for compute devices stored in its database. We performed verification of running the two applications on up to 512 and 1024 processes respectively on a large cluster from Academic Computer Center in Gdansk demonstrating a high degree of accuracy between simulated and measured results.

show abstract

Section: Related Workmentioning

confidence: 99%

Modeling energy consumption of parallel applications

Czarnul

Kuchta

Rościszewski

et al. 2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

show abstract

“…Modeling is a useful vehicle to approximate power, energy, and temperature on systems that lack direct measurement capabilities. These models employ performance information to estimate the desired parameters [17,18]. In addition, modeling is essential to predict power and energy on future systems.…”

Section: Related Workmentioning

confidence: 99%

Program optimizations: The interplay between power, performance, and energy

et al. 2016

View full text Add to dashboard Cite

Practical considerations for future supercomputer designs will impose limits on both instantaneous power consumption and total energy consumption. Working within these constraints while providing the maximum possible performance, application developers will need to optimize their code for speed alongside power and energy concerns. This paper analyzes the effectiveness of several code optimizations including loop fusion, data structure transformations, and global allocations. A per component measurement and analysis of different architectures is performed, enabling the examination of code optimizations on different compute subsystems. Using an explicit hydrodynamics proxy application from the U.S. Department of Energy, LULESH, we show how code optimizations impact different computational phases of the simulation. This provides insight for simulation developers into the best optimizations to use during particular simulation compute phases when optimizing code for future supercomputing platforms. We examine and contrast both x86 and Blue Gene architectures with respect to these optimizations.

show abstract

“…Subramaniam et al built a regression model for the power and performance of scientific applications and used this model to optimize energy efficiency [37]. Tiwari et al developed CPU and DIMM power and energy models of three widely used HPC kernels by training artificial neural networks [43]. While these models provide good estimation of power and/or performance metrics, they cannot capture the dynamic, complicated powerperformance interactions exhibiting in large-scale systems.…”

Section: Related Workmentioning

confidence: 99%

Quantitative modeling of power performance tradeoffs on extreme scale systems

Zhou

Wallace

Papka

et al. 2015

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

h i g h l i g h t s• A colored Petri net was developed for tradeoff analysis of power and performance.• Trace based validation demonstrated that the model is highly accurate and scalable.• The model was used to analyze different power capping methods on petascale systems. a b s t r a c t As high performance computing (HPC) continues to grow in scale and complexity, energy becomes a critical constraint in the race to exascale computing. The days of ''performance at all cost'' are coming to an end. While performance is still a major objective, future HPC will have to deliver desired performance under the energy constraint. Among various power management methods, power capping is a widely used approach. Unfortunately, the impact of power capping on system performance, user jobs, and powerperformance efficiency are not well studied due to many interfering factors imposed by system workload and configurations. To fully understand power management in extreme scale systems with a fixed power budget, we introduce a power-performance modeling tool named PuPPET (Power Performance PETri net). Unlike the traditional performance modeling approaches such as analytical methods or trace-based simulators, we explore a new approach -colored Petri nets -for the design of PuPPET. PuPPET is fast and extensible for navigating through different configurations. More importantly, it can scale to hundreds of thousands of processor cores and at the same time provide high levels of modeling accuracy. We validate PuPPET by using system traces (i.e., workload log and power data) collected from the production 48-rack IBM Blue Gene/Q supercomputer at Argonne National Laboratory. Our trace-based validation demonstrates that PuPPET is capable of modeling the dynamic execution of parallel jobs on the machine by providing an accurate approximation of energy consumption. In addition, we present two case studies of using PuPPET to study power-performance tradeoffs on petascale systems.

show abstract

Modeling Power and Energy Usage of HPC Kernels

Cited by 46 publications

References 23 publications

Modeling energy consumption of parallel applications

Modeling energy consumption of parallel applications

Program optimizations: The interplay between power, performance, and energy

Quantitative modeling of power performance tradeoffs on extreme scale systems

Contact Info

Product

Resources

About