An On-line Algorithm for Fair-Share Node Allocations in a Cluster

Amar, Lior; Barak, Amnon; Levy, Ely; Okun, Michael

doi:10.1109/ccgrid.2007.22

Cited by 16 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For better support of GPU computing on clusters, it is important to develop MOSIX-like algorithms for dynamic resource management, load-balancing, task priorities, fairshare and a method for choosing the "best" device [10], [22]. It is also important to add network optimizations, such as support for direct exchange of boundaries among different devices in the scatter-gather mechanism.…”

Section: Discussionmentioning

confidence: 99%

A package for OpenCL based heterogeneous computing on clusters with many GPU devices

Barak

Ben-Nun

Levy

et al. 2010

2010 IEEE International Conference on Cluster Computing Workshops and Posters (CLUSTER WORKSHOPS)

Self Cite

View full text Add to dashboard Cite

Abstract-Heterogeneous systems provide new opportunities to increase the performance of parallel applications on clusters with CPU and GPU architectures. Currently, applications that utilize GPU devices run their device-executable code on local devices in their respective hosting-nodes. This paper presents a package for running OpenMP, C++ and unmodified OpenCL applications on clusters with many GPU devices. This Many GPUs Package (MGP) includes an implementation of the OpenCL specifications and extensions of the OpenMP API that allow applications on one hosting-node to transparently utilize cluster-wide devices (CPUs and/or GPUs). MGP provides means for reducing the complexity of programming and running parallel applications on clusters, including scheduling based on task dependencies and buffer management. The paper presents MGP and the performance of its internals.

show abstract

Section: Discussionmentioning

confidence: 99%

A package for OpenCL based heterogeneous computing on clusters with many GPU devices

Barak

Ben-Nun

Levy

et al. 2010

2010 IEEE International Conference on Cluster Computing Workshops and Posters (CLUSTER WORKSHOPS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Feasible share of resources (fes i ), similarly to the concept presented in [4] for the fair-share policy, determines the exact number of resources that should be allocated to each user, considering his/her demand and owner share, and the unused nodes by users that have d i < a i . Its value is determined by:…”

Section: The Owner-share Enforcement Policymentioning

confidence: 99%

“…Amar and others [4] use the concept of fair-share to build an algorithm for node allocation in a cluster, where they used preemption in a space-sharing approach. Their algorithm is concerned with addition or subtraction of nodes into the system.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

The Owner Share Scheduler for a Distributed System

Falavinha

Manacero

Livny

et al. 2009

2009 International Conference on Parallel Processing Workshops

View full text Add to dashboard Cite

Abstract-In large distributed systems, where shared resources are owned by distinct entities, there is a need to reflect resource ownership in resource allocation. An appropriate resource management system should guarantee that resource's owners have access to a share of resources proportional to the share they provide. In order to achieve that some policies can be used for revoking access to resources currently used by other users. In this paper, a scheduling policy based in the concept of distributed ownership is introduced called Owner Share Enforcement Policy (OSEP). OSEP goal is to guarantee that owner do not have their jobs postponed for longer periods of time. We evaluate the results achieved with the application of this policy using metrics that describe policy violation, loss of capacity, policy cost and user satisfaction in environments with and without job checkpointing. We also evaluate and compare the OSEP policy with the Fair-Share policy, and from these results it is possible to capture the trade-offs from different ways to achieve fairness based on the user satisfaction.

show abstract

“…For example, in [10] the authors presented a distributed proportional-share scheduler for a cluster that uses the DBB for obtaining information about the current usage of the cluster. Additionally, in a system that supports load-balancing, an overloaded node can use the information obtained from the local vector to migrate processes or virtual machines to less-loaded nodes, or to migrate processes from slower to faster nodes, subject to the relative loads of the nodes.…”

mentioning

confidence: 99%

Randomized gossip algorithms for maintaining a distributed bulletin board with guaranteed age properties

Amar

Barak

Drezner

et al. 2009

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SUMMARYScalable computer systems, including clusters and multi-cluster grids, require routine exchange of information about the state of system-wide resources among their nodes. Gossip-based algorithms are popular for providing such information services due to their simplicity, fault tolerance and low communication overhead. This paper presents a randomized gossip algorithm for maintaining a distributed bulletin board among the nodes of a scalable computer system. In this algorithm each node routinely disseminates its most recently acquired information while maintaining a snapshot of the other nodes' states. The paper provides analytical approximations for the expected average age, the age distribution and the expected maximal age for the acquired information at each node. We confirm our results by measurements of the performance of the algorithm on a multi-cluster campus grid with 256 nodes and by simulations of configurations with up to 2048 nodes. The paper then presents practical enhancements of the algorithm, which makes it more suitable for a real system. Such enhancements include using fixed-size messages, reducing the number of messages sent to inactive nodes and supporting urgent information. The enhanced algorithm guarantees the age properties of the information at each node in the configurations with an arbitrary number of inactive nodes. It is being used in our campus grid for resource discovery, for dynamic assignment of processes to the best available nodes, for load-balancing and for on-line monitoring.

show abstract

An On-line Algorithm for Fair-Share Node Allocations in a Cluster

Cited by 16 publications

References 7 publications

A package for OpenCL based heterogeneous computing on clusters with many GPU devices

A package for OpenCL based heterogeneous computing on clusters with many GPU devices

The Owner Share Scheduler for a Distributed System

Randomized gossip algorithms for maintaining a distributed bulletin board with guaranteed age properties

Contact Info

Product

Resources

About