Home Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty

Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty

Session recording and materials

Presentation slides

Join Red Hat Research for the next Research Days event, “Endure: A Robust Tuning Paradigm for LSM Trees Under Workload Uncertainty ” on February 8, 2023 from 11AM to 12:30PM EST (5PM CET, 6PM IST).  

Tuning database performance in cloud systems with uncertain workloads is notoriously difficult.  In this event, you will learn about a new robust tuning methodology that can consistently outperform standard tunings by up to five times. Join us and bring your questions to discuss how this methodology could help improve your cloud database deployments. Andy HuynhManos Athanassoulis, and Evimaria Terzi, all of Boston University, will be our speakers and Josh Berkus, Red Hat, will lead the conversation.

Log-Structured Merge trees (LSM trees) are increasingly used as the storage engines behind several data systems, frequently deployed in the cloud. Similar to other database architectures, LSM trees consider information about the expected workload (e.g., reads vs. writes, point vs. range queries) to optimize their performance via tuning. However, operating in a shared infrastructure like the cloud comes with workload uncertainty due to the fast-evolving nature of modern applications. Systems with static tuning discount the variability of such hybrid workloads and hence provide inconsistent and overall suboptimal performance. In this talk we introduce Endure – a new paradigm for tuning LSM trees in the presence of workload uncertainty. We introduce the robust problem and show how it differs from the traditional tuning problem. Then we will discuss how to define uncertainty and use this definition to recommend tunings that perform on average better under the presence of a changing workload. We show the potential of robust tunings by implementing Endure in RockDB, a popular LSM engine, and analyze the performance benefits of using robust tunings.

Andy Huynh, Boston University
Manos Athanassoulis, Boston University
Evimaria Terzi, Boston University

Conversation Leader:
Josh Berkus, Red Hat

Visit project pages for the Robust LSM-Trees Under Workload Uncertainty and Robust Data Systems Tuning projects to learn more.

The recording and materials will be available following the talk.

Learn more about Red Hat Research Days Events and watch the recordings from previous events at research.redhat.com/research-talks


Feb 08 2023


11:00 am - 12:30 pm

Local Time

  • Timezone: America/New_York
  • Date: Feb 08 2023
  • Time: 11:00 am - 12:30 pm



Submit a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.