A General "Power-of-d" Dispatching Framework for Heterogeneous Systems
Jazeem Abdul Jaleel, Sherwin Doroudi, Kristen Gardner, Alexander, Wickeham

TL;DR
This paper develops a flexible framework for dispatching in heterogeneous large-scale systems, leveraging server speed info at multiple decision points to optimize response times.
Contribution
It introduces a general heterogeneity-aware power-of-d dispatching framework that separates querying and assignment policies, enabling analysis and optimization in heterogeneous settings.
Findings
Optimized policies improve mean response time.
Heuristic policies perform well in simulations.
Framework accommodates diverse server speeds and decision rules.
Abstract
Intelligent dispatching is crucial to obtaining low response times in large-scale systems. One common scalable dispatching paradigm is the ``power-of-,'' in which the dispatcher queries servers at random and assigns the job to a server based only on the state of the queried servers. The bulk of power-of- policies studied in the literature assume that the system is homogeneous, meaning that all servers have the same speed; meanwhile real-world systems often exhibit server speed heterogeneity. This paper introduces a general framework for describing and analyzing heterogeneity-aware power-of- policies. The key idea behind our framework is that dispatching policies can make use of server speed information at two decision points: when choosing which servers to query, and when assigning a job to one of those servers. Our framework explicitly separates the dispatching…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsOptimization and Search Problems · Cloud Computing and Resource Management · Complexity and Algorithms in Graphs
