MPI and communication---Adaptive, transparent frequency and voltage scaling of communication phases in MPI programs

Buyya

Seventh IEEE International Symposium on Cluster Computing and the Grid (CCGrid '07)

2007

252

155

Section: Related Workmentioning

confidence: 99%

Power Aware Scheduling of Bag-of-Tasks Applications with Deadline Constraints on DVS-enabled Clusters

Buyya

Seventh IEEE International Symposium on Cluster Computing and the Grid (CCGrid '07)

2007

252

155

2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing

“…Now we discuss the relationship between e i,j .cost and PE's operating frequency. Figure 2 shows the energy consumption and communication cost as processor frequency varies for four common MPI calls when different size of data are transferred among PEs [18]. From the experiment results we can see energy can be saved up to 31% with at most 5% communication time increase.…”

Section: Parallel Task Modelmentioning

confidence: 99%

“…• Some work applies DVFS during the communication phases of high performance computing, for example MPI [17], [18].…”

Section: B Energy Reduction Via Dvfs Techniquementioning

confidence: 99%

Towards Energy Aware Scheduling for Precedence Constrained Parallel Tasks in a Cluster with DVFS

Wang

Laszewski

Dayal

et al. 2010

187

108

Abstract-Reducing energy consumption for high end computing can bring various benefits such as, reduce operating costs, increase system reliability, and environment respect. This paper aims to develop scheduling heuristics and to present application experience for reducing power consumption of parallel tasks in a cluster with the Dynamic Voltage Frequency Scaling (DVFS) technique. In this paper, formal models are presented for precedence-constrained parallel tasks, DVFS enabled clusters, and energy consumption. This paper studies the slack time for non-critical jobs, extends their execution time and reduces the energy consumption without increasing the task's execution time as a whole. Additionally, Green Service Level Agreement is also considered in this paper. By increasing task execution time within an affordable limit, this paper develops scheduling heuristics to reduce energy consumption of a tasks execution and discusses the relationship between energy consumption and task execution time. Models and scheduling heuristics are examined with a simulation study. Test results justify the design and implementation of proposed energy aware scheduling heuristics in the paper.

2012 IEEE 24th International Symposium on Computer Architecture and High Performance Computing

“…Our work also differs from previous works in that we use partial phase recognition instead of phase prediction. Online recognition of communication phases in MPI application was investigated by Lim et al in [7]. Once a communication phase is recognized, authors apply CPU DVFS to save energy.…”

Section: Related Workmentioning

confidence: 99%

Beyond CPU Frequency Scaling for a Fine-grained Energy Control of HPC Systems

Chetsa

Lefèvre

Pierson

et al. 2012

Abstract-Modern high performance computing subsystems (HPC) -including processor, network, memory, and IOare provided with power management mechanisms. These include dynamic speed scaling and dynamic resource sleeping. Understanding the behavioral patterns of high performance computing systems at runtime can lead to a multitude of optimization opportunities including controlling and limiting their energy usage.In this paper, we present a general purpose methodology for optimizing energy performance of HPC systems considering processor, disk and network. We rely on the concept of execution vector along with a partial phase recognition technique for on-the-fly dynamic management without any a priori knowledge of the workload. We demonstrate the effectiveness of our management policy under two real-life workloads. Experimental results show that our management policy in comparison with baseline unmanaged execution saves up to 24% of energy with less than 4% performance overhead for our real-life workloads.