PCS: Predictive Component-level Scheduling for Reducing Tail Latency in Cloud Online Services

TL;DR

PCS is a predictive, component-level scheduling framework designed to reduce tail latency in cloud online services by modeling component performance and dynamically allocating resources.

Contribution

It introduces an analytical performance model for predicting component and service latency, enabling adaptive scheduling to mitigate interference effects.

Findings

Reduces component tail latency by 67.05% on average.

Decreases overall service latency by 64.16%.

Outperforms existing tail latency reduction techniques.

Abstract

Modern latency-critical online services often rely on composing results from a large number of server components. Hence the tail latency (e.g. the 99th percentile of response time), rather than the average, of these components determines the overall service performance. When hosted on a cloud environment, the components of a service typically co-locate with short batch jobs to increase machine utilizations, and share and contend resources such as caches and I/O bandwidths with them. The highly dynamic nature of batch jobs in terms of their workload types and input sizes causes continuously changing performance interference to individual components, hence leading to their latency variability and high tail latency. However, existing techniques either ignore such fine-grained component latency variability when managing service performance, or rely on executing redundant requests to reduce the tail latency, which adversely deteriorate the service performance when load gets heavier. In this paper, we propose PCS, a predictive and component-level scheduling framework to reduce tail latency for large-scale, parallel online services. It uses an analytical performance model to simultaneously predict the component latency and the overall service performance on different nodes. Based on the predicted performance, the scheduler identifies straggling components and conducts near-optimal component-node allocations to adapt to the changing performance interferences from batch jobs. We demonstrate that, using realistic workloads, the proposed scheduler reduces the component tail latency by an average of 67.05\% and the average overall service latency by 64.16\% compared with the state-of-the-art techniques on reducing tail latency.