Filters








4 Hits in 2.6 sec

Policy driven heterogeneous resource co-allocation with Gangmatching

R. Raman, M. Livny, M. Solomon
High Performance Distributed Computing, 2003. Proceedings. 12th IEEE International Symposium on  
However, Matchmaking is limited by its purely bilateral formalism of matching a single customer with a single resource, precluding more advanced resource management services such as co-allocation.  ...  Conventional resource management methods that rely on static models of resource allocation policy and behavior fail to address these challenges.  ...  Distributed ownership introduces policy heterogeneity: Each resource may have its own idiosyncratic allocation policy.  ... 
doi:10.1109/hpdc.2003.1210018 dblp:conf/hpdc/RamanLS03 fatcat:bnovh3prirb7nhzg52mvtwkbdq

Leveraging Public Resource Pools to Improve the Service Compliances of Computing Utilities [chapter]

Shah Asaduzzaman, Muthucumaru Maheswaran
2004 Lecture Notes in Computer Science  
One of the major objectives of computing utilities is to maximize their net profit while maintaining customer loyalty in accordance with the service level agreements (SLAs).  ...  Defining the SLAs conservatively might be one easy way to achieve SLA compliance, but this results in underutilization of resources and loss of revenue in turn.  ...  The Condor project [11] focuses on harvesting unused resources from heterogeneous public machines, but their resource management mainly emphasizes on discovery and co-allocation of resources through  ... 
doi:10.1007/978-3-540-30474-6_29 fatcat:2xgcm6hlbzaezbgokeq4la75ry

Utilizing unreliable public resources for higher profit and better SLA compliance in computing utilities

Shah Asaduzzaman, Muthucumaru Maheswaran
2006 Journal of Parallel and Distributed Computing  
However, by tuning the SLA parameters conservatively the computing utility might under utilize its resources with a resultant loss of revenue.  ...  SLAs are satisfied with very high probability.  ...  The Condor project [30] focuses on harvesting unused resources from heterogeneous public machines, but their resource management mainly emphasizes on discovery and co-allocation of resources through  ... 
doi:10.1016/j.jpdc.2006.01.005 fatcat:mjvl6gziqzgotltwugdnhcvwzu

An evaluation of the close-to-files processor and data co-allocation policy in multiclusters

H.H. Mohamed, D.H.J. Epema
2004 IEEE International Conference on Cluster Computing (IEEE Cat. No.04EX935)  
In multicluster systems, and more generally, in grids, johs may require co-allocation, i.e., the simultaneous allocation of resources such as pmcessors and inputfiles in multiple clusters.  ...  Here, we extend this work with an analysis of the p e r f o m n c e in a real testhed of our pmtotype Processor and Data Co-Allocator with the Close-to-Files (CF) job-placement algorirhm CF tries to place  ...  Our results show that the combination of the CF policy and file replication is very beneficial, that the system utilization with a mixed workload of co-allocated and local jobs can be driven to about 80%  ... 
doi:10.1109/clustr.2004.1392626 dblp:conf/cluster/MohamedE04 fatcat:n64gtefmbfg5xpndhq47verj4e