DMRlib: Easy-coding and efficient resource management for job malleability
Process malleability has proved to have a highly positive impact on the resource utilization and global productivity in data centers compared with the conventional static resource allocation policy. However, the non-negligible additional development effort this solution imposes has constrained its a...
| Autores: | , , , |
|---|---|
| Formato: | artículo |
| Fecha de publicación: | 2020 |
| País: | España |
| Recursos: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/329704 |
| Acesso em linha: | https://hdl.handle.net/2117/329704 https://dx.doi.org/10.1109/TC.2020.3022933 |
| Access Level: | acceso abierto |
| Palavra-chave: | High performance computing Data centers Processes Reconfiguration MPI malleability Job Elastic Resize Dynamic Reallocation of Resources Productivity-Aware Computation Càlcul intensiu (Informàtica) Àrees temàtiques de la UPC::Informàtica::Enginyeria del software |
| Resumo: | Process malleability has proved to have a highly positive impact on the resource utilization and global productivity in data centers compared with the conventional static resource allocation policy. However, the non-negligible additional development effort this solution imposes has constrained its adoption by the scientific programming community. In this work, we present DMRlib, a library designed to offer the global advantages of process malleability while providing a minimalist MPI-like syntax. The library includes a series of predefined communication patterns that greatly ease the development of malleable applications. In addition, we deploy several scenarios to demonstrate the positive impact of process malleability featuring different scalability patterns. Concretely, we study two job submission modes (rigid and moldable) in order to identify the best-case scenarios for malleability using metrics such as resource allocation rate, completed jobs per second, and energy consumption. The experiments prove that our elastic approach may improve global throughput by a factor higher than 3x compared to the traditional workloads of non-malleable jobs. |
|---|