Managing local disk write latency in Jira Data Center

Local disk write latency indicator measures the time taken to write data to the local disk on each Jira Data Center node. This metric is crucial for understanding the performance of individual nodes within your Jira Data Center cluster.

How does high local disk write latency affect Jira Data Center performance

High local disk write latency can significantly impact your Jira Data Center instance in several ways:

• Slower Lucene search operations (JQL queries)

• Increased time for issue creation and updates

• Delayed node-to-node cache communications

• Extended reindexing times (full, background, and project reindexes)

• Overall degradation of Jira performance and user experience

Consistently high Local Disk Write Latency on one or more nodes can lead to performance inconsistencies across your Jira Data Center cluster.

What's the recommendation?

You should aim to keep the local disk write latency below 1ms for optimal performance. If you're consistently seeing latencies above 1ms, consider the following actions:

1. Check for environmental factors

2. Verify platform compatibility

3. Optimize disk configuration

4. Diagnose runtime issues

Check for environmental factors

Various environmental factors can contribute to high Local Disk Write Latency.

Common environmental factors to investigate:

1. Anti-virus software: Ensure it's not interfering with Jira's disk operations

2. Remote or shared drives: Verify that local home directories are on local disks

3. Virtualized environments: Check for any VM-specific disk performance issues

4. Ongoing maintenance tasks: Look for disk defragmentation or backup processes

5. File system encryption or compression: Assess their impact on write performance

Steps to mitigate environmental factors:

1. Configure anti-virus software to exclude Jira directories from real-time scanning

2. Ensure local home directories are on high-performance local storage

3. Optimize VM settings for disk I/O performance

4. Schedule maintenance tasks during off-peak hours

5. Evaluate the necessity of file system encryption or compression

See our guide on Interpreting infrastructure metrics for in-product diagnostics.

Verify platform compatibility

Ensuring your local disk setup is compatible with Jira Data Center requirements is essential for optimal performance.

Key compatibility checks:

1. Verify that you're using supported operating systems and file systems

2. Ensure local home directories are not on NFS mounts (not supported for Jira DC)

3. Check that your storage solution meets Jira's I/O performance requirements

Refer to the Supported platforms for Jira Data Center documentation

If you identify compatibility issues:

1. Plan to migrate to supported platforms if necessary

2. Consider upgrading storage hardware to meet performance requirements

3. Consult with your infrastructure team to ensure optimal configuration

Optimize disk configuration

Proper disk configuration is critical for maintaining low Local Disk Write Latency in Jira Data Center.

Optimization steps:

1. Ensure local home directories are on SSDs or high-performance HDDs

2. Optimize file system settings for write performance

3. Adjust I/O scheduler settings for optimal performance

4. Consider separating Jira's local home directory from the OS and application directories

Refer to the Installing Jira Data Center guide for more details

Diagnose runtime issues

Java Flight Recorder (JFR) can provide valuable insights into JVM behavior that might be contributing to local disk latency.

Using Java Flight Recorder:

1. Enable JFR on all Jira Data Center nodes

2. Collect JFR recordings during periods of high Local Disk Write Latency

3. Analyze the recordings to identify any JVM-related issues affecting disk performance

See our guide on Diagnosing runtime issues using the Java Flight Recorder

Interpreting JFR results:

• Identify any threads that are blocking I/O operations

• Check for patterns in disk access that might indicate inefficient data access or caching

Additional considerations for Jira Data Center:

1. Monitor local disk write Latency across all nodes to identify any inconsistencies

2. Ensure that all nodes have similar hardware specifications and configurations

3. Consider the impact of cluster-wide operations (e.g., reindexing) on local disk performance

4. Regularly review and optimize Jira's database and index configurations to reduce disk I/O