Analyzing Resource Trade-offs in Hardware Overprovisioned Supercomputers
| Autor: | Thang Cao, Masatsugu Ueda, Daniel Ellsworth, Masaaki Kondo, Ryuichi Sakamoto, Tapasya Patki, Koji Inoue, Barry Rountree, Martin Schulz |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Job scheduler
Computer science business.industry 020209 energy Node (networking) Control (management) 02 engineering and technology computer.software_genre Scheduling (computing) Constraint (information theory) Resource (project management) 0202 electrical engineering electronic engineering information engineering Key (cryptography) Resource management business Throughput (business) computer Computer hardware |
| Zdroj: | IPDPS |
| Popis: | Hardware overprovisioned systems have recently been proposed as a viable alternative for a power-efficient design of next-generation supercomputers. A key challenge for such systems is to determine the degree of overprovisioning, which refers to the number of extra nodes that need to be installed under a given power constraint. In this paper, we first show that the degree of overprovisioning depends on dynamic parameters, such as the job mix as well as the global power constraint, and that static decisions can result in limited system throughput. We then study an exhaustive combination of adaptive resource management strategies that span three job scheduling algorithms, four power capping techniques, and three node boot-up mechanisms to understand the trade-off space involved. We then draw conclusions about how these strategies can adaptively control the degree of overprovisioning and analyze their impact on job throughput and power utilization. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|