Shekharrajak commented on PR #3903:
URL: 
https://github.com/apache/datafusion-comet/pull/3903#issuecomment-4193553081

   @andygrove I have noted down the benchmark result here : 
https://github.com/apache/datafusion-comet/pull/3904 
   
   this the table I got : 
   
   | Metric | main (baseline) | After single spill file  | Delta |
   |--------|----------------|---------|-------|
   | **avg write time** | 1.376s | 1.113s | **-19.1%** |
   | **min write time** | 1.349s | 1.106s | **-18.0%** |
   | **max write time** | 1.415s | 1.126s | **-20.4%** |
   | **throughput** | 3,633,169 rows/s | 4,490,434 rows/s | **+23.6%** |
   | write time (I/O) | 0.113s (8.2%) | 0.037s (3.3%) | **-67.3%** |
   | encode time | 0.745s (54.1%) | 0.726s (65.2%) | -2.5% (noise) |
   | repart time | 0.041s (3.0%) | 0.041s (3.7%) | 0% |
   | output size | 226.20 MiB | 4.38 MiB | writes compressed data during the 
spill phase, so the final output file contains already-compressed data |
   | spill count | 10 | 10 | identical |
   | data size (in-memory) | 440.36 MiB | 440.36 MiB | identical |


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to