ScyllaDB University Live | Free Virtual Training Event
Learn more
ScyllaDB Documentation Logo Documentation
  • Server
  • Cloud
  • Tools
    • ScyllaDB Manager
    • ScyllaDB Monitoring Stack
    • ScyllaDB Operator
  • Drivers
    • CQL Drivers
    • DynamoDB Drivers
  • Resources
    • ScyllaDB University
    • Community Forum
    • Tutorials
Download
ScyllaDB Docs ScyllaDB Monitoring Using Scylla Monitoring Stack Scylla Monitoring Stack Advisor Compaction takes lots of memory and CPU

Caution

You're viewing documentation for a previous version of ScyllaDB Monitoring. Switch to the latest stable version.

Compaction takes lots of memory and CPUΒΆ

Scylla runs compaction periodically as a background process. While running compaction is important, there are situations that the compaction takes too much CPU. In those cases the compaction would impact the overall system performance.

When facing this problem, you can statically limit the compaction shares with one of two options:

  1. Change scylla.yml to have compaction_static_shares: 100

or

  1. Start scylla with --compaction-static-shares 100

Was this page helpful?

PREVIOUS
Some queries use reverse order
NEXT
Some operation failed due to unsatisfied consistency level
  • Create an issue
  • Edit this page
ScyllaDB Monitoring
  • 3.10
    • 4.9
    • 4.8
    • 4.7
    • 4.6
    • 4.5
    • 4.4
    • 4.3
    • 4.2
    • 4.1
    • 4.0
    • 3.10
    • 3.9
    • 3.8
    • 3.7
    • 3.6
    • 3.5
  • Introduction
  • Download and Install
    • Install
    • The start-all.sh script
    • Deploy without Docker
    • Docker Compose
    • System Recommendations
    • Using Thanos
  • User Guide
    • CQL Optimization Dashboard
    • Advisor
      • Some queries use ALLOW FILTERING
      • Some queries use Consistency Level: ALL
      • Some queries use Consistency Level: ANY
      • Some queries are not token-aware
      • Some SELECT queries are non-paged
      • Some queries are non-prepared
      • Some queries use reverse order
      • Compaction takes lots of memory and CPU
      • Some operation failed due to unsatisfied consistency level
      • I/O Errors can indicate a node with a faulty disk
      • Some operations failed on the replica side
      • CQL queries are not balanced among shards
      • Prepared statements cache eviction
      • System Overload
  • Procedures
    • Alert Manager
      • Alerting
    • Adding and Modifying Dashboards
    • Upgrade Guide
  • Upgrade
    • Monitoring 3.x to 3.y
    • Monitoring 2.x to 3.y
    • Monitoring 2.x to 2.y
    • Monitoring 1.x to 2.x
  • Troubleshooting
    • Troubleshooting
    • Troubleshooting Guide for Scylla Manager and Scylla Monitor Integration
  • Reference
    • Support Matrix
    • Interfaces
  • GitHub Project
Docs Tutorials University Contact Us About Us
© 2025, ScyllaDB. All rights reserved. | Terms of Service | Privacy Policy | ScyllaDB, and ScyllaDB Cloud, are registered trademarks of ScyllaDB, Inc.
Last updated on 04 May 2025.
Powered by Sphinx 7.4.7 & ScyllaDB Theme 1.8.6