Evaluating adaptive and predictive power management strategies for optimizing visualization performance on supercomputers
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Univ. of Oregon, Eugene, OR (United States)
Power is becoming an increasingly scarce resource on the next generation of supercomputers, and should be used wisely to improve overall performance. One strategy for improving power usage is hardware overprovisioning, i.e., systems with more nodes than can be run at full power simultaneously without exceeding the system-wide power limit. With this study, we compare two strategies for allocating power throughout an overprovisioned system – adaptation and prediction – in the context of visualization workloads. While adaptation has been suitable for workloads with more regular execution behaviors, it may not be as suitable on visualization workloads, since they can have variable execution behaviors. This study considers a total of 104 experiments, which vary the rendering workload, power budget, allocation strategy, and node concurrency, including tests processing data sets up to 1 billion cells and using up to 18,432 cores across 512 nodes. Overall, we find that prediction is a superior strategy for this use case, improving performance up to 27% compared to an adaptive strategy.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE
- Grant/Contract Number:
- AC52-07NA27344; LLNL-JRNL-782698
- OSTI ID:
- 1834504
- Alternate ID(s):
- OSTI ID: 1894380
- Report Number(s):
- LLNL-JRNL-782698; 977642
- Journal Information:
- Parallel Computing, Vol. 104-105; ISSN 0167-8191
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Priority research directions for in situ data management: Enabling scientific discovery from diverse data sources
|
journal | March 2020 |
Power Tuning HPC Jobs on Power-Constrained Systems
|
conference | September 2016 |
Just-in-time dynamic voltage scaling: Exploiting inter-node slack to save energy in MPI programs
|
journal | September 2008 |
On the Greenness of In-Situ and Post-Processing Visualization Pipelines
|
conference | May 2015 |
Evaluation of In-Situ Analysis Strategies at Scale for Power Efficiency and Scalability
|
conference | May 2016 |
VTK-m: Accelerating the Visualization Toolkit for Massively Threaded Architectures
|
journal | May 2016 |
Trapped Capacity: Scheduling under a Power Cap to Maximize Machine-Room Throughput
|
conference | November 2014 |
Cinema image-based in situ analysis and visualization of MPAS-ocean simulations
|
journal | July 2016 |
Energy-Aware Framework for Virtual Machine Consolidation in Cloud Computing
|
conference | November 2013 |
Dynamic power sharing for higher job throughput
|
conference | November 2015 |
CPU MISER: A Performance-Directed, Run-Time System for Power-Aware Clusters
|
conference | September 2007 |
Exploring tradeoffs between power and performance for a scientific visualization algorithm
|
conference | October 2015 |
In Situ Visualization for Computational Science
|
journal | November 2019 |
In Situ Methods, Infrastructures, and Applications on High Performance Computing Platforms
|
journal | June 2016 |
Performance Modeling of In Situ Rendering
|
conference | November 2016 |
Similar Records
Communication Characterization and Optimization of Applications Using Topology-Aware Task Mapping on Large Supercomputers
Communication Characterization and Optimization of Applications Using Topology-Aware Task Mapping on Large Supercomputers