Title: "Jockey: Guaranteed Job Latency in Data Parallel Clusters" [pdf]
Venue: EuroSys'12
Abstract:
Data processing frameworks such as MapReduce and Dryad are used today in business environments where customers expect guaranteed performance. To date, however, these systems are not capable of providing guarantees on job latency because scheduling policies are based on fair-sharing, and operator seek high cluster use through statistical multiplexing and over-subscription. With Jockey, we provide latency SLOs for data parallel jobs written in SCOPE. Jockey precomputes statistics using a simulator that captures the job's complex internal dependencies, accurately and effectively predicting the remaining run time at different resource allocations and in different stages of the job. Our control policy monitor a job's performance, and dynamically adjusts resource allocation in the shared cluster in order to maximize the job's economic utility while minimizing its impact on the rest of the cluster. In our experiments in Microsoft's production Cosmos clusters, Jockey meets the specified job SLOs and responds to changes in cluster conditions.