Surprising results on task assignment in server farms with high-variability workloads

Authors:
Mor Harchol-Balter;Alan Scheller-Wolf;Andrew R. Young
Affiliations:
Carnegie Mellon University, Pittsburgh, PA, USA;Carnegie Mellon University, Pittsburgh, PA, USA;N/A, Pittsburgh, PA, USA
Venue:
Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems
Year:
2009

Citing 22
Cited 9

Exploiting process lifetime distributions for dynamic load balancing

Proceedings of the 1996 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Self-similarity in World Wide Web traffic: evidence and possible causes

Proceedings of the 1996 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Generating representative Web workloads for network and server performance evaluation

SIGMETRICS '98/PERFORMANCE '98 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Load-sensitive routing of long-lived IP flows

Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
On choosing a task assignment policy for a distributed server system

Journal of Parallel and Distributed Computing - Special issue on software support for distributed computing
Task assignment with unknown duration

Journal of the ACM (JACM)
EQUILOAD: a load balancing policy for clustered web servers

Performance Evaluation
Delay moments for FIFO GI/GI/s queues

Queueing Systems: Theory and Applications
New bounds for expected delay in FIFO GI/GI/c queues

Queueing Systems: Theory and Applications
Further delay moment results for FIFO multiserver queues

Queueing Systems: Theory and Applications
The impact of a heavy-tailed service-time distribution upon the M/GI/s waiting-time distribution

Queueing Systems: Theory and Applications
Waiting-Time Asymptotics for the M/G/2 Queue with Heterogeneous Servers

Queueing Systems: Theory and Applications
Task Assignment Strategy for Overloaded Systems

ISCC '03 Proceedings of the Eighth IEEE International Symposium on Computers and Communications
Evaluation of Task Assignment Policies for Supercomputing Servers: The Case for Load Unbalancing and Fairness

Cluster Computing
A least flow-time first load sharing approach for distributed server farm

Journal of Parallel and Distributed Computing
Heavy Tails in Multi-Server Queue

Queueing Systems: Theory and Applications
Systems with multiple servers under heavy-tailed workloads

Performance Evaluation - Performance 2005
Optimal state-free, size-aware dispatching for heterogeneous M/G/-type systems

Performance Evaluation - Performance 2005
Structural interpretation and derivation of necessary and sufficient conditions for delay moments in FIFO multiserver queues

Queueing Systems: Theory and Applications
Task assignment with work-conserving migration

Parallel Computing
Allocation of Service Time in a Multiserver System

Management Science
Comparing job allocation schemes where service demand is unknown

Journal of Computer and System Sciences

Analysis of SITA policies

Performance Evaluation
Scheduling policies for single-hop networks with heavy-tailed traffic

Allerton'09 Proceedings of the 47th annual Allerton conference on Communication, control, and computing
Why segregating short jobs from long jobs under high variability is not always a win

Allerton'09 Proceedings of the 47th annual Allerton conference on Communication, control, and computing
Optimal allocation of servers and processing time in a load balancing system

Computers and Operations Research
Review: Task assignment policies in distributed server systems: A survey

Journal of Network and Computer Applications
Minimizing slowdown in heterogeneous size-aware dispatching systems

Proceedings of the 12th ACM SIGMETRICS/PERFORMANCE joint international conference on Measurement and Modeling of Computer Systems
Queue-length asymptotics for generalized max-weight scheduling in the presence of heavy-tailed traffic

IEEE/ACM Transactions on Networking (TON)
Steady-State Analysis for Multiserver Queues Under Size Interval Task Assignment in the Quality-Driven Regime

Mathematics of Operations Research
Lookahead actions in dispatching to parallel queues

Performance Evaluation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper investigates the performance of task assignment policies for server farms, as the variability of job sizes (service demands) approaches infinity. Our results reveal that some common wisdoms regarding task assignment are flawed. The Size-Interval-Task-Assignment policy (SITA), which assigns each server a unique size range, was heretofore thought of by some as the panacea for dealing with high-variability job-size distributions. We show SITA to be inferior to the much simpler greedy policy, Least-Work-Left (LWL), for certain common job-size distributions, including many modal, hyperexponential, and Pareto distributions. We also define regimes where SITA's performance is superior, and prove simple closed-form bounds on its performance for the above-mentioned distributions.