Power-aware MPI task aggregation prediction for high-end computing systems

Liu, Dong; Nikolopoulos, Dimitrios S.; Cameron, Kirk W.; Supinski, Bronis R. de; Schulz, Martin

doi:10.1109/ipdps.2010.5470464

Cited by 22 publications

(16 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practice, pack scheduling is really useful as shown by recent results. Li et al [15] propose a framework to predict the energy and performance impacts of power-aware MPI task aggregation. Frachtenberg et al [9] show that system utilization can be improved through their schemes to co-schedule jobs based on their load-balancing requirements and inter-processor communication patterns.…”

Section: Related Workmentioning

confidence: 99%

Co-scheduling algorithms for high-throughput workload execution

Aupy

Shantharam

Benoît

et al. 2015

J Sched

View full text Add to dashboard Cite

International audienceThis paper investigates co-scheduling algorithms for processing a set of parallel applications. Instead of executing each application one by one, using a maximum degree of parallelism for each of them, we aim at scheduling several applications concurrently. We partition the original application set into a series of packs, which are executed one by one. A pack comprises several applications, each of them with an assigned number of processors, with the constraint that the total number of processors assigned within a pack does not exceed the maximum number of available processors. The objective is to determine a partition into packs, and an assignment of processors to applications, that minimize the sum of the execution times of the packs. We thoroughly study the complexity of this optimization problem , and propose several heuristics that exhibit very good performance on a variety of workloads, whose application execution times model profiles of parallel scientific codes. We show that co-scheduling leads to faster workload completion time (40% improvement on average over traditional scheduling) and to faster response times (50% improvement). Hence co-scheduling increases system throughput and saves energy, leading to significant benefits from both the user and system perspectives

show abstract

Section: Related Workmentioning

confidence: 99%

Co-scheduling algorithms for high-throughput workload execution

Aupy

Shantharam

Benoît

et al. 2015

J Sched

View full text Add to dashboard Cite

show abstract

“…In [1], Y. Ma et al explain how energy consumption can be reduced by efficient task clustering with task duplication. The paper [12] discusses how task aggregation can contribute to energy savings. Our technique allocates resources considering OOP as a parameter along with resource requirement to reduce energy consumption.…”

Section: Related Workmentioning

confidence: 99%

Energy efficient rescheduling algorithm for High Performance Computing

Chauhan

Parveen

Saurav

et al. 2015

2015 National Conference on Parallel Computing Technologies (PARCOMPTECH)

View full text Add to dashboard Cite

With the augmentation of High Performance Computing (HPC) system and power consumption pattern, energy optimization has become an important concern. Several surveys indicate that the energy utilized in computation and communication within a HPC system contributes considerably to their operational costs. The proposed energy efficient rescheduling algorithm aims to reduce energy consumption in HPC. This algorithm is based on energy efficient rescheduling of a job which considers the optimal operating point (OOP) i.e. voltage and frequency along with resource matching constraint to achieve the target performance. Performance -Energy-Time (PET) matrix for every application is built and stored in knowledge base (KB) in order to devise its OOP. The devised OOP gives minimal energy consumption at target execution environment. The algorithm is useful for energy optimization for cluster environment. Earlier works exploited process migration technique to address load balancing aspect. The proposed algorithm exploits process migration and reschedules the job dynamically with its OOP in context of energy reduction.

show abstract

“…In [16] Li takes a modeling approach to investigate task aggregation to reduce energy consumption by reducing the number of nodes. Li uses AMG along with some NPB MPI benchmarks, one of the few efforts that use a real HPC applications in their analysis.…”

Section: Related Workmentioning

confidence: 99%

“…Their work predicts the performance energy tradeoff up to 1024 cores (128 nodes). In both [15] and [16] the System G supercomputer at Virginia Tech is used. System G consists of 324 nodes.…”

Section: Related Workmentioning

confidence: 99%

Measuring and tuning energy efficiency on large scale high performance computing platforms.

Laros¹

2011

View full text Add to dashboard Cite

Recognition of the importance of power in the field of High Performance Computing, whether it be as an obstacle, expense or design consideration, has never been greater and more pervasive. While research has been conducted on many related aspects, there is a stark absence of work focused on large scale High Performance Computing. Part of the reason is the lack of measurement capability currently available on small or large platforms. Typically, research is conducted using coarse methods of measurement such as inserting a power meter between the power source and the platform, or fine grained measurements using custom instrumented boards (with obvious limitations in scale). To collect the measurements necessary to analyze real scientific computing applications at large scale, an in-situ measurement capability must exist on a large scale capability class platform.In response to this challenge, we exploit the unique power measurement capabilities of the Cray XT architecture to gain an understanding of power use and the effects of tuning. We apply these capabilities at the operating system level by deterministically halting cores when idle. At the application level, we gain an understanding of the power requirements of a range of important DOE/NNSA production scientific computing applications running at large scale (thousands of nodes), while simultaneously collecting current and voltage measurements on the hosting nodes. We examine the effects of both CPU and network bandwidth tuning and demonstrate energy savings opportunities of up to 39% with little or no impact on run-time performance. Capturing scale effects in our 3 experimental results was key. Our results provide strong evidence that next generation large-scale platforms should not only approach CPU frequency scaling differently, but could also benefit from the capability to tune other platform components, such as the network, to achieve energy efficient performance. 5Acknowledgments I would like to acknowledge the following people who have contributed in various ways, all of which significant, to the successful completion of this work: Dr.

show abstract

Power-aware MPI task aggregation prediction for high-end computing systems

Cited by 22 publications

References 26 publications

Co-scheduling algorithms for high-throughput workload execution

Co-scheduling algorithms for high-throughput workload execution

Energy efficient rescheduling algorithm for High Performance Computing

Measuring and tuning energy efficiency on large scale high performance computing platforms.

Contact Info

Product

Resources

About