Deciding which queue to join: Some counterexamples
Operations Research
Wide area traffic: the failure of Poisson modeling
IEEE/ACM Transactions on Networking (TON)
Exploiting process lifetime distributions for dynamic load balancing
ACM Transactions on Computer Systems (TOCS)
Long-lasting transient conditions in simulations with heavy-tailed workloads
Proceedings of the 29th conference on Winter simulation
Self-similarity in World Wide Web traffic: evidence and possible causes
IEEE/ACM Transactions on Networking (TON)
Load-balancing heuristics and process behavior
SIGMETRICS '86/PERFORMANCE '86 Proceedings of the 1986 ACM SIGMETRICS joint international conference on Computer performance modelling, measurement and evaluation
Analysis of the Early Workload on the Cornell Theory Center IBM SP2
Analysis of the Early Workload on the Cornell Theory Center IBM SP2
On Choosing a Task Assignment Policy for a Distributed Server System
On Choosing a Task Assignment Policy for a Distributed Server System
Directory Based Composite Routing and Scheduling for Dynamic Multimedia Environments
IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium
Metrics for Parallel Job Scheduling and Their Convergence
JSSPP '01 Revised Papers from the 7th International Workshop on Job Scheduling Strategies for Parallel Processing
Formal Models of Heavy-Tailed Behavior in Combinatorial Search
CP '01 Proceedings of the 7th International Conference on Principles and Practice of Constraint Programming
Parallel physics-inspired waterflow particle mechanics algorithm for load rebalancing
Computer Networks: The International Journal of Computer and Telecommunications Networking
Teams organization and performance analysis in autonomous human-robot teams
Proceedings of the 10th Performance Metrics for Intelligent Systems Workshop
Hi-index | 0.00 |
We consider a distributed server system model and ask which policy should be used for assigning tasks to hosts. In our model each host processes tasks in First-Come-First-Serve order and the task's service demand is known in advance. We consider four task assignment policies commonly proposed for such distributed server systems: Round-Robin, Random, Size-Based, in which all tasks within a give size range are assigned to a particular host, and Dynamic-Least-Work-Remaining, in which a task is assigned to the host with the least outstanding work. Our goal is to understand the influence of task size variability on the decision of which task assignment policy is best. We find that no one of the above task assignment policies is best and that the answer depends critically on the variability in the task size distribution. In particular we find that when the task sizes are not highly variable, the Dynamic policy is preferable. However when task sizes show the degree of variability more characteristic of empirically measured computer workloads, the Size-Based policy is the best choice. We use the resulting observations to argue in favor of a specific size-based policy, SITA-E, that can outperform the Dynamic policy by almost 2 orders of magnitude and can outperform other task assignment policies by many orders of magnitude, under a realistic task size distribution.