Market-based autonomous resource and application management in private clouds

Handle URI:
http://hdl.handle.net/10754/623160
Title:
Market-based autonomous resource and application management in private clouds
Authors:
Costache, Stefania; Kortas, Samuel; Morin, Christine; Parlavantzas, Nikos
Abstract:
High Performance Computing (HPC) clouds need to be efficiently shared between selfish tenants having applications with different resource requirements and Service Level Objectives (SLOs). The main difficulty relies on providing concurrent resource access to such tenants while maximizing the resource utilization. To overcome this challenge, we propose Merkat, a market-based SLO-driven cloud platform. Merkat relies on a market-based model specifically designed for on-demand fine-grain resource allocation to maximize resource utilization and it uses a combination of currency distribution and dynamic resource pricing to ensure proper resource distribution among tenants. To meet the tenant’s SLO, Merkat uses autonomous controllers, which apply adaptation policies that: (i) dynamically tune the application’s provisioned CPU and memory per virtual machine in contention periods, or (ii) dynamically change the number of virtual machines. Our evaluation with simulation and on the Grid’5000 testbed shows that Merkat provides flexible support for different application types and SLOs and good tenant satisfaction compared to existing centralized systems, while the infrastructure resource utilization is improved.
KAUST Department:
KAUST, 4700-Thuwal, Jeddah Zip Code 23955-6900, Saudi Arabia.
Citation:
Costache S, Kortas S, Morin C, Parlavantzas N (2017) Market-based autonomous resource and application management in private clouds. Journal of Parallel and Distributed Computing 100: 85–102. Available: http://dx.doi.org/10.1016/j.jpdc.2016.10.003.
Publisher:
Elsevier BV
Journal:
Journal of Parallel and Distributed Computing
Issue Date:
12-Oct-2016
DOI:
10.1016/j.jpdc.2016.10.003
Type:
Article
ISSN:
0743-7315
Sponsors:
This work was done while the first author was a Ph.D. student at INRIA and EDF R&D and was supported by ANRT through the CIFRE sponsorship No. 0332/2010. Experiments presented in this paper were carried out using the Grid’5000 experimental testbed, being developed under the INRIA ALADDIN development action with support from CNRS, RENATER and several Universities as well as other funding bodies (see https://www.grid5000.fr).
Additional Links:
http://www.sciencedirect.com/science/article/pii/S0743731516301204
Appears in Collections:
Articles

Full metadata record

DC FieldValue Language
dc.contributor.authorCostache, Stefaniaen
dc.contributor.authorKortas, Samuelen
dc.contributor.authorMorin, Christineen
dc.contributor.authorParlavantzas, Nikosen
dc.date.accessioned2017-04-13T11:50:58Z-
dc.date.available2017-04-13T11:50:58Z-
dc.date.issued2016-10-12en
dc.identifier.citationCostache S, Kortas S, Morin C, Parlavantzas N (2017) Market-based autonomous resource and application management in private clouds. Journal of Parallel and Distributed Computing 100: 85–102. Available: http://dx.doi.org/10.1016/j.jpdc.2016.10.003.en
dc.identifier.issn0743-7315en
dc.identifier.doi10.1016/j.jpdc.2016.10.003en
dc.identifier.urihttp://hdl.handle.net/10754/623160-
dc.description.abstractHigh Performance Computing (HPC) clouds need to be efficiently shared between selfish tenants having applications with different resource requirements and Service Level Objectives (SLOs). The main difficulty relies on providing concurrent resource access to such tenants while maximizing the resource utilization. To overcome this challenge, we propose Merkat, a market-based SLO-driven cloud platform. Merkat relies on a market-based model specifically designed for on-demand fine-grain resource allocation to maximize resource utilization and it uses a combination of currency distribution and dynamic resource pricing to ensure proper resource distribution among tenants. To meet the tenant’s SLO, Merkat uses autonomous controllers, which apply adaptation policies that: (i) dynamically tune the application’s provisioned CPU and memory per virtual machine in contention periods, or (ii) dynamically change the number of virtual machines. Our evaluation with simulation and on the Grid’5000 testbed shows that Merkat provides flexible support for different application types and SLOs and good tenant satisfaction compared to existing centralized systems, while the infrastructure resource utilization is improved.en
dc.description.sponsorshipThis work was done while the first author was a Ph.D. student at INRIA and EDF R&D and was supported by ANRT through the CIFRE sponsorship No. 0332/2010. Experiments presented in this paper were carried out using the Grid’5000 experimental testbed, being developed under the INRIA ALADDIN development action with support from CNRS, RENATER and several Universities as well as other funding bodies (see https://www.grid5000.fr).en
dc.publisherElsevier BVen
dc.relation.urlhttp://www.sciencedirect.com/science/article/pii/S0743731516301204en
dc.subjectResource managementen
dc.subjectCloud computingen
dc.subjectElastic scalingen
dc.subjectMarket mechanismsen
dc.subjectService Level Objectiveen
dc.subjectHPCen
dc.titleMarket-based autonomous resource and application management in private cloudsen
dc.typeArticleen
dc.contributor.departmentKAUST, 4700-Thuwal, Jeddah Zip Code 23955-6900, Saudi Arabia.en
dc.identifier.journalJournal of Parallel and Distributed Computingen
dc.contributor.institutionEDF R&D, 1 av. du General De Gaule, BP 408, 92141 Clamart cedex, Franceen
dc.contributor.institutionINRIA, Campus de Beaulieu, 35042 Rennes cedex, Franceen
dc.contributor.institutionIRISA, Campus de Beaulieu, 35042 Rennes cedex, Franceen
kaust.authorKortas, Samuelen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.