Managing clusters of geographically distributed high-performance computers

Brune, Matthias; Gehring, Jörn; Keller, Axel; Reinefeld, Alexander

doi:10.1002/(sici)1096-9128(19991225)11:15<887::aid-cpe459>3.0.co;2-j

Cited by 19 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In both cases, jobs are routed (migrated) towards the clusters' LRMSs. The Hierarchical 2-level metascheduler is the most encountered variant [8,11]. Distributed meta-scheduler: similarly to the independent clusters architecture, each cluster has its LRMS, and jobs arrive at the individual clusters' LRMSs.…”

Section: Independent Clusters: (Included For Completeness)mentioning

confidence: 99%

“…Completely decentralized systems can be scalable and fault-tolerant, but they can be much less efficient than their hierarchical alternatives. While many solutions have already been proposed [4,8,9,11,31,33,34], none has so far managed to achieve acceptance in the grid world, in part because they have yet to prove that they can yield significant benefits when managing typical grid workloads.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inter-Operating Grids Through Delegated MatchMaking

Iosup

Tannenbaum

Farrellee

et al. 2008

Scientific Programming

View full text Add to dashboard Cite

Abstract. The grid vision of a single computing utility has yet to materialize: while many grids with thousands of processors each exist, most work in isolation. An important obstacle for the effective and efficient inter-operation of grids is the problem of resource selection. In this paper we propose a solution to this problem that combines the hierarchical and decentralized approaches for interconnecting grids. In our solution, a hierarchy of grid sites is augmented with peer-to-peer connections between sites under the same administrative control. To operate this architecture, we employ the key concept of delegated matchmaking, which temporarily binds resources from remote sites to the local environment. With trace-based simulations we evaluate our solution under various infrastructural and load conditions, and we show that it outperforms other approaches to inter-operating grids. Specifically, we show that delegated matchmaking achieves up to 60% more goodput and completes 26% more jobs than its best alternative.

show abstract

Section: Independent Clusters: (Included For Completeness)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Inter-Operating Grids Through Delegated MatchMaking

Iosup

Tannenbaum

Farrellee

et al. 2008

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…On the one hand requirements from the application are specified. Advanced reservations and information about future resource usage are mandatory to guarantee that a multi-site [2,6] application starts synchronously. Only a minority of the available resource management systems provide these features.…”

Section: Introductionmentioning

confidence: 99%

Scheduling in HPC Resource Management Systems: Queuing vs. Planning

Hovestadt

Kao

Keller

et al. 2003

Lecture Notes in Computer Science

Self Cite

112

View full text Add to dashboard Cite

Abstract. Nearly all existing HPC systems are operated by resource management systems based on the queuing approach. With the increasing acceptance of grid middleware like Globus, new requirements for the underlying local resource management systems arise. Features like advanced reservation or quality of service are needed to implement high level functions like co-allocation. However it is difficult to realize these features with a resource management system based on the queuing concept since it considers only the present resource usage. In this paper we present an approach which closes this gap. By assigning start times to each resource request, a complete schedule is planned. Advanced reservations are now easily possible. Based on this planning approach functions like diffuse requests, automatic duration extension, or service level agreements are described. We think they are useful to increase the usability, acceptance and performance of HPC machines. In the second part of this paper we present a planning based resource management system which already covers some of the mentioned features.

show abstract

“…The support of multi-site parallel execution [12,13,14,15,16] on a computational grid has been examined in previous works, concerning the execution of a job in parallel at different sites. Under the condition of a limited communication overhead, the results from our previous work [10] and from [1,3,4] all showed that multi-site parallel execution can improve the overall average response time.…”

Section: Related Workmentioning

confidence: 99%

Towards Feasible and Effective Load Sharing in a Heterogeneous Computational Grid

Huang

Shih

Chung

Advances in Grid and Pervasive Computing

View full text Add to dashboard Cite

Abstract.A grid has to provide strong incentive for participating sites to join and stay in it. Participating sites are concerned with the performance improvement brought by the gird for the jobs of their own local user communities. Feasible and effective load sharing is key to fulfilling such a concern. This paper explores the load-sharing policies concerning feasibility and heterogeneity on computational grids. Several job scheduling and processor allocation policies are proposed and evaluated through a series of simulations using workloads derived from publicly available trace data. The simulation results indicate that the proposed job scheduling and processor allocation policies are feasible and effective in achieving performance improvement on a heterogeneous computational grid.

show abstract

Managing clusters of geographically distributed high-performance computers

Cited by 19 publications

References 28 publications

Inter-Operating Grids Through Delegated MatchMaking

Inter-Operating Grids Through Delegated MatchMaking

Scheduling in HPC Resource Management Systems: Queuing vs. Planning

Towards Feasible and Effective Load Sharing in a Heterogeneous Computational Grid

Contact Info

Product

Resources

About