Decoupling computation and data scheduling in distributed data-intensive applications

By Kavitha Ranganathan, Ian T Foster. Proceedings 11th IEEE International Symposium on High Performance Distributed Computing, 2002

540 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/fd1030baf61086977d357d53c98cbaf4bff777a6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

A peer-to-peer approach to resource location in grid environments (2002, background): “Existing Gridsimulators are specialized forcertain services, such asscheduling [LMC03] or data replication [ RF02 ].”
The Grid Economy (2005, background): “While scheduling of bandwidth or storage-intensive applications is currently a hot topic in the Grid research community [34], [38], most of them give concentrate on the job turnaround time and give less importance to the limitations of the network and storage resources such as bandwidth and storage ”
Grid'5000: a large scale and highly reconfigurable grid experimental testbed (2005, background): “ChicSim [8] and HyperSim [9] are other simulators close to the presented ones.”
Scheduling Algorithms for Grid Computing: State of the Art and Open Problems (2006, background): “In [86], four simple dynamic computation scheduling algorithms (namely, JobRandom which selects a random site for computation; JobLeastLoaded which selects the least loaded site; JobDataPresent which always selects a site where the required data is already there, and JobLocal which always runs a job”
Cloud auto-scaling with deadline and budget constraints (2010, background): “(1) Some works [98][99] discussed the ideas of duplicating the data among geographically distributed computing sites to reduce data transfers.”
Versatile, scalable, and accurate simulation of distributed applications and platforms (2014, unknown): “Some were made available to the community but proved short-lived, such as OptorSim [8] (shown in the table) or ChicSim [68].”
On Death, Taxes, and the Convergence of Peer-to-Peer and Grid Computing (2003, background): “as reliable and scalable management of large job pools, distributed scheduling [40], replica location [11], and discovery [16].”
SimGrid: A Generic Framework for Large-Scale Distributed Experiments (2008, background): “Most of these tools only intend to be used by their own developers, and even when they are released to the public they target a very specific community.”
Mapping Abstract Complex Workflows onto Grid Environments (2003, background): “In [39] significant performance improvement is achieved when scheduling is performed according to data availability while also using a dynamic replication strategy.”
Resource allocation in the grid using reinforcement learning (2004, background): “Ranganathan and Foster[ 21 ] describe a scheduling framework for a system that is very similar to ours: large number of users that submit jobs to one of the large number of sites.”