A comparison of local and gang scheduling on a Beowulf cluster

Authors:
P. Strazdins;J. Uhlmann
Affiliations:
Dept. of Comput. Sci., Australian Nat. Univ., Acton, ACT, Australia;Dept. of Comput. Sci., Australian Nat. Univ., Acton, ACT, Australia
Venue:
CLUSTER '04 Proceedings of the 2004 IEEE International Conference on Cluster Computing
Year:
2004

Citing 0
Cited 7

Performance Comparison of Coscheduling Algorithms for Non-Dedicated Clusters Through a Generic Framework

International Journal of High Performance Computing Applications
The Impact of Critical Sporadic Jobs on Gang Scheduling Performance in Distributed Systems

Simulation
Time-sharing parallel applications through performance-targeted feedback-controlled real-time scheduling

Cluster Computing
Multi-site scheduling with multiple job reservations and forecasting methods

ISPA'06 Proceedings of the 4th international conference on Parallel and Distributed Processing and Applications
A Bee Colony based optimization approach for simultaneous job scheduling and data replication in grid environments

Computers and Operations Research
Static and dynamic job scheduling with communication aware policy in cluster computing

Computers and Electrical Engineering
Hopfield neural network for simultaneous job scheduling and data replication in grids

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Gang scheduling and related techniques are widely believed to be necessary for efficient job scheduling on distributed memory parallel computers. This is because they minimize context switching overheads and permit the parallel job currently running to progress at the fastest possible rate. However, in the case of cluster computers, and particularly those with COTS networks, these benefits can be outweighed in the multiple jobs time-sharing context by the loss the ability to utilize the CPU for other jobs when the current job is waiting for messages. Experiments on a Linux Beowulf cluster with 100 Mb fast Ethernet switches are made comparing the SCore buddy-based gang scheduling with local scheduling (provided by the Linux 2.4 kernel with MPI implemented over TCP/IP). Results for communication-intensive numerical applications on 16 nodes reveal that gang scheduling results in 'slowdowns ' up to a factor of two greater for 8 simultaneous jobs. This phenomenon is not due to any deficiencies in SCore but due to the relative costs of context switching versus message overhead, and we expect similar results holds for any gang scheduling implementation. A performance analysis of local scheduling indicates that cache pollution due to context switching is more significant than the direct context switching overhead on the applications studied. When this is taken into account, local scheduling behaviour comes close to achieving ideal slowdowns for finer-grained computations such as Linpack. The performance models also indicate that similar trends are to be expected for clusters with faster networks.