Skip to Main content Skip to Navigation

Merkat: Market-based Autonomous Application and Resource Management in the Cloud

Abstract : Organizations owning HPC infrastructures are facing difficulties in managing their infrastructures. These difficulties come from the need to provide concurrent resource access to applications with different resource requirements while considering that users might have different performance objectives, or Service Level Objectives (SLOs) for executing them. To address these challenges this paper proposes a market-based SLO-driven cloud platform. This platform relies on a market-based model to allocate resources to applications while taking advantage of cloud flexibility to maximize resource utilization. The combination of currency distribution and dynamic resource pricing ensures fair resource distribution. In the same time, autonomous controllers apply adaptation policies to scale the application resource demand according to user SLOs. The adaptation policies can: (i) dynamically tune the amount of CPU and memory provisioned for the virtual machines in contention periods; (ii) dynamically change the number of virtual machines. We evaluated this proposed platform on the Grid'5000 testbed. Results show that: (i) the platform provides flexible support for different application types and different SLOs; (ii) the platform is capable to provide good user satisfaction achieving acceptable performance degradation compared to existing centralized solutions.
Complete list of metadata

Cited literature [28 references]  Display  Hide  Download
Contributor : Stefania Victoria Costache Connect in order to contact the contributor
Submitted on : Friday, August 9, 2013 - 11:40:35 AM
Last modification on : Tuesday, October 19, 2021 - 11:58:53 PM
Long-term archiving on: : Wednesday, April 5, 2017 - 8:29:06 PM


Files produced by the author(s)


  • HAL Id : hal-00850862, version 1


Stefania Victoria Costache, Nikos Parlavantzas, Christine Morin, Samuel Kortas. Merkat: Market-based Autonomous Application and Resource Management in the Cloud. [Research Report] RR-8343, INRIA. 2013. ⟨hal-00850862⟩



Les métriques sont temporairement indisponibles