- Exploiting Spatio-Temporal Tradeoffs for Energy-Aware MapReduce in the Cloud
Michael Cardosa, Aameek Singh, Himabindu Pucha and Abhishek Chandra.
To appear in IEEE Transactions on Computers.
Description: We describe a unique spatio-temporal
tradeoff for achieving energy efficiency for MapReduce jobs in virtualized
environments.
- STEAMEngine: Driving MapReduce Provisioning in the Cloud
Michael Cardosa, Piyush Narang, Abhishek Chandra, Himabindu Pucha and Aameek
Singh.
In Proceedings of the 18th Annual International Conference on High
Performance Computing (HiPC'11), Bangalore, India, December 2011.
Description: We present
intelligent provisioning algorithms that optimize for user-side
performance and provider-side energy metrics.
- Exploiting Spatio-Temporal Tradeoffs for Energy Efficient MapReduce in the Cloud
Michael Cardosa, Aameek Singh, Himabindu Pucha and Abhishek Chandra
To appear in the 4th IEEE International Conference on Cloud Computing (CLOUD'11), July 2011, Washington DC, USA.
Description: We describe a unique spatio-temporal tradeoff for achieving energy efficiency for MapReduce jobs in virtualized environments.
- Exploring MapReduce Efficiency with Highly-Distributed Data [pdf]
Michael Cardosa, Chenyu Wang, Anshuman Nangia, Abhishek Chandra and Jon Weissman
To appear in the 2nd International Workshop on MapReduce and its Applications (MAPREDUCE'11), June 2011, San Jose, CA, USA.
Description: We explore several architectural approaches to constructing MapReduce clusters when the resources and data are highly distributed.
- STEAMEngine: Driving MapReduce Provisioning in the Cloud
Michael Cardosa, Piyush Narang, Abhishek Chandra, Himabindu Pucha and Aameek Singh
Poster at the 9th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2010), Vancouver, BC, Canada, October 2010.
Technical Report TR10-023, Department of Computer Science and Engineering, University of Minnesota, September 2010.
Tech Report
- Resource Bundles: Using Aggregation for Statistical Large-Scale Resource Discovery and Management
Michael Cardosa and Abhishek Chandra
In IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol. 21, No. 8, August 2010.
Description: Using trace-driven simulations and data analysis of a PlanetLab trace, we show that resource bundles are ideally suited for identifying group-level characteristics. We present an adaptive algorithm to parameterize the bundling algorithm.
- HiDRA: Statistical Multi-dimensional Resource Discovery for Large-scale Systems
Michael Cardosa and Abhishek Chandra
In Proceedings of the 17th IEEE International Workshop on Quality of Service (IWQoS'09), Charleston, SC, July 2009.
Description: We present HiDRA, a scalable resource discovery technique providing statistical guarantees for resource requirements spanning multiple dimensions simultaneously. Through trace analysis and a PlanetLab implementation, we show that HiDRA performs nearly as well as a fully-informed algorithm and is a feasible, low-cost approach to statistical resource discovery.
- Shares and Utilities based Power Consolidation in Virtualized Server Environments
Michael Cardosa, Madhukar Korupolu and Aameek Singh
In Proceedings of the 11th IEEE International Symposium on Integrated Network Management (IM 2009), New York, NY, June 2009.
Description: We present a novel suite of techniques for placement and power consolidation of VMs in data centers. We confirm the end-to-end validity of our approach and demonstrate that our final candidate algorithm provides a practical solution for administrators.
- Resource Bundles: Using Aggregation for Statistical Wide-Area Resource Discovery and Allocation
Michael Cardosa and Abhishek Chandra
In Proceedings of the 28th International Conference on Distributed Computing Systems (ICDCS 2008), Beijing, China, June 2008.
Description: We propose the notion of a resource bundle, employ two complementary techniques to provide statistical guarantees for resource capacities, and use clustering-based techniques for hierarchical aggregation and scalability.
|
