Skip to main content

Troubleshooting Performance Issues in Coalesce

Check Your Local Environment​

Start by ruling out issues with your local setup. These are the most common sources of slowness and are the quickest to fix.

  • Use a Supported Browser: Confirm that you are using Google Chrome. Other browsers are not officially supported and may cause unexpected behavior.
  • Check Your Network Connection: As a browser-based application, Coalesce requires a strong and stable network connection to perform optimally. Check if other websites are loading quickly or run a network speed test.
  • Review Security Software: Corporate security software can sometimes block components of the Coalesce service, leading to performance degradation. Work with your security or IT team to ensure that Coalesce is allowed and not being blocked.
  • Free Up System Resources: If your computer's memory (RAM) is nearly full, you may see an impact on Coalesce's performance. Close unnecessary applications and browser tabs to free up resources.

Pipeline Design and Optimization​

An inefficient pipeline design can significantly impact run times, regardless of the underlying data platform. Review your graph for common performance issues.

  • Review your data pipeline Operations like GROUP BY, DISTINCT, and ORDER BY are computationally expensive and can slow down your jobs. Use these only when necessary.
  • Use persistent staging tables. Instead of creating complex, nested views, use persistent staging tables to break up long-running operations and improve parallel execution.
  • Break down complex pipelines. Large, monolithic pipelines can be difficult to manage and optimize. Break complex pipelines into more manageable, separable chunks.
  • Use incremental loading. For large datasets, use proper incremental loading strategies to process only new or changed data, rather than performing full table scans with every run.

Snowflake-Specific Optimizations​

If you are using Snowflake, these optimizations may help.

  • Check warehouse sizing. Make sure your Snowflake warehouse is the right size for your jobs. If you see jobs waiting in line or taking too long, try using a larger warehouse.
  • Check query details. Use Snowflake's query history to look for performance bottlenecks like large table scans or disk spilling, which indicates the warehouse is too small for the operation.
  • Consider clustering keys or search optimization. For very large tables, implementing a clustering key or using Snowflake's search optimization service can dramatically improve query performance.
  • Check how many jobs are running at once. If multiple jobs are running at the same time, you may see them waiting in line. Check your warehouse setup to make sure it can handle all the jobs running at once.

If you are still experiencing slowness after checking these items, we encourage you to open a request with Coalesce Support for further investigation.