Themis: Economy-Based Automatic Resource Scaling for Cloud Systems

Abstract : High performance computing (HPC) infrastructures are used to execute increasingly complex scientific applications with time-varying resource requirements. A key challenge for infrastructure providers is to distribute resources to such applications so that application performance objectives are met while guaranteeing a high infrastructure utilization. Most of existing solutions provide limited support for meeting application objectives and can lead to inefficient resource usage. In this paper we propose Themis, a system that uses an economic-based approach to automatically allocate resources to the applications that need them the most. Themis relies on a proportional-share auction that ensures fair differentiation between applications while maximizing the resource utilization. To support meeting application performance objectives, Themis provides generic scaling policies based on feedback control loops. These policies adapt the resource demand to the current infrastructure conditions and can be extended for different application types. We have evaluated the performance of the system through simulations, and the results show that Themis can effectively meet application performance objectives while optimizing infrastructure's utilization.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/hal-00698583
Contributor : Stefania Victoria Costache <>
Submitted on : Saturday, May 4, 2013 - 7:22:15 PM
Last modification on : Thursday, November 15, 2018 - 11:57:45 AM
Long-term archiving on : Friday, March 31, 2017 - 8:33:51 AM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00698583, version 1

Citation

Stefania Victoria Costache, Nikos Parlavantzas, Christine Morin, Samuel Kortas. Themis: Economy-Based Automatic Resource Scaling for Cloud Systems. 14th IEEE International Conference on High Performance Computing and Communications (HPCC 2012), Jun 2012, Liverpool, United Kingdom. ⟨hal-00698583⟩

Share

Metrics

Record views

699

Files downloads

451