Compiler-Directed Energy-Time Tradeoff in MPI Programs on DVS-Enabled Parallel Systems
نویسندگان
چکیده
Although parallel systems with high peak performance have been exciting, high peak performance often means high power consumption. In this paper, power-aware parallel systems are investigated, where each node can make dynamic voltage scaling (DVS). Based on the characteristics of communication and memory access in MPI programs, a compiler is used to automatically form communication and computation regions, and to optimally assign frequency and voltage to the regions. Frequency and voltage of each node are dynamically adjusted, and energy consumption is minimized within the limit of performance loss. The results from simulations and experiments show that compiler-directed energy-time tradeoff can save 20~40% energy consumption with less than 5% performance loss.
منابع مشابه
Energy Efficient Real-Time Scheduling in Distributed Systems
Battery powered real-time systems have been widely used in many applications. As the quantity and the functional complexity of battery powered devices continue to increase, energy efficient design of such devices has become important. Also these realtime systems have to concurrently perform a multitude of complex tasks with strict time constraints. Thus, minimizing power consumption and extendi...
متن کاملParallel computing using MPI and OpenMP on self-configured platform, UMZHPC.
Parallel computing is a topic of interest for a broad scientific community since it facilitates many time-consuming algorithms in different application domains.In this paper, we introduce a novel platform for parallel computing by using MPI and OpenMP programming languages based on set of networked PCs. UMZHPC is a free Linux-based parallel computing infrastructure that has been developed to cr...
متن کاملCompiler-Directed Dynamic Voltage Scaling for Memory-Bound Applications
This paper presents the design and implementation of a compiler algorithm that effectively reduces the energy usage of memory-bound applications via dynamic voltage scaling (DVS). The algorithm identifies program regions where the CPU can be slowed down with negligible performance penalty. It is implemented as a source-to-source level transformation using the SUIF2 compiler infrastructure. Phys...
متن کاملAN AUTOMATED APPROACH TO IMPROVING COMMUNICATION-COMPUTATION OVERLAP IN CLUSTERS by
Parallel clusters have become common platforms for programmers to achieve desired runtime performance for applications with high processing demands. Unfortunately, scaling these applications to larger numbers of CPUs for even higher performance gains often fails because the communication overhead increases at a similar rate. The compiler optimization research group at the University of Delaware...
متن کاملImplementing an Open64-based Tool for Improving the Performance of MPI Programs
While MPI parallel programming has become the primary approach to achieving performance gains in cluster computing, the communication overhead inherent in a cluster environment continues to be a major obstacle. A promising approach to improve performance is the use of computationcommunication overlapping, which is enabled by communication libraries that utilize Remote Direct Memory Access (RDMA...
متن کامل