Open shivangi24 opened 3 weeks ago
Would you locate the stage on SQL UI ? Then we can see some basic metrics.
It's likely related to scan's fallback and possibly caused by either slow scan or slow R2C. Could check on the metrics to identify.
Attaching the screenshots of stage which took longer along with its metrics.
Hi @pratham76 thanks.
Although can we view the query on SQL UI? There is a SQL / DataFrame
tab on the RHS of the tabs.
Backend
VL (Velox)
Bug description
We are currently working on integrating Gluten into our WatsonX.Data's Spark environment. However, after enabling Gluten and running the TPCH benchmark at the 100G scale, we are not observing the performance improvements as claimed in the Gluten repository. Specifically, we are seeing only a 10-12% improvement, whereas a 2x improvement is expected.
Here are the details of our environment:
We have experimented with various configurations, but the performance gain has not exceeded 10-12% across all 22 queries. We have attached a graph showing the performance comparison between runs with and without Gluten.
Adding spark events for single query - Q6 f2b74f64-bdfe-42ba-a6f7-ad81028cb2d7_events.zip cc: @deepashreeraghu @majetideepak
Spark version
Spark-3.4.x
Spark configurations
Ran with 2 executors of (6*24G)
System information
No response
Relevant logs
We observed that one stage took significantly longer to complete. Could you please investigate the cause of the delay?