Re: [PR] Perf: fast CursorValues compare for StringViewArray using inline_key_โ€ฆ [datafusion]

Wed, 02 Jul 2025 00:29:11 -0700 


zhuqi-lucas commented on PR #16630:
URL: https://github.com/apache/datafusion/pull/16630#issuecomment-3026764953

   > ๐Ÿค–: Benchmark completed
   > 
   > Details
   > 
   > ```
   > Comparing HEAD and fast_sort_with_inlined_fast_key
   > --------------------
   > Benchmark sort_tpch.json
   > --------------------
   > 
โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ณโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ณโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ณโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”“
   > โ”ƒ Query        โ”ƒ       HEAD โ”ƒ fast_sort_with_inlined_fast_key โ”ƒ        
Change โ”ƒ
   > 
โ”กโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ•‡โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ•‡โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ•‡โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ฉ
   > โ”‚ Q1           โ”‚  325.73 ms โ”‚                       343.28 ms โ”‚  1.05x 
slower โ”‚
   > โ”‚ Q2           โ”‚  323.40 ms โ”‚                       296.73 ms โ”‚ +1.09x 
faster โ”‚
   > โ”‚ Q3           โ”‚ 1094.10 ms โ”‚                      1061.63 ms โ”‚     no 
change โ”‚
   > โ”‚ Q4           โ”‚  413.50 ms โ”‚                       422.10 ms โ”‚     no 
change โ”‚
   > โ”‚ Q5           โ”‚  396.88 ms โ”‚                       403.57 ms โ”‚     no 
change โ”‚
   > โ”‚ Q6           โ”‚  433.31 ms โ”‚                       437.50 ms โ”‚     no 
change โ”‚
   > โ”‚ Q7           โ”‚  792.53 ms โ”‚                       791.69 ms โ”‚     no 
change โ”‚
   > โ”‚ Q8           โ”‚  665.29 ms โ”‚                       684.66 ms โ”‚     no 
change โ”‚
   > โ”‚ Q9           โ”‚  700.98 ms โ”‚                       727.78 ms โ”‚     no 
change โ”‚
   > โ”‚ Q10          โ”‚ 1027.52 ms โ”‚                      1042.44 ms โ”‚     no 
change โ”‚
   > โ”‚ Q11          โ”‚  579.11 ms โ”‚                       549.83 ms โ”‚ +1.05x 
faster โ”‚
   > 
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
   > โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ณโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”“
   > โ”ƒ Benchmark Summary                              โ”ƒ           โ”ƒ
   > โ”กโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ•‡โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”ฉ
   > โ”‚ Total Time (HEAD)                              โ”‚ 6752.35ms โ”‚
   > โ”‚ Total Time (fast_sort_with_inlined_fast_key)   โ”‚ 6761.22ms โ”‚
   > โ”‚ Average Time (HEAD)                            โ”‚  613.85ms โ”‚
   > โ”‚ Average Time (fast_sort_with_inlined_fast_key) โ”‚  614.66ms โ”‚
   > โ”‚ Queries Faster                                 โ”‚         2 โ”‚
   > โ”‚ Queries Slower                                 โ”‚         1 โ”‚
   > โ”‚ Queries with No Change                         โ”‚         8 โ”‚
   > โ”‚ Queries with Failure                           โ”‚         0 โ”‚
   > โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
   > ```
   
   The Q1 regression seems noise, because it does not have utf8view sort key, 
the l_linenumber is Integer:
   ```rust
         r#"
           SELECT l_linenumber, l_partkey
           FROM lineitem
           ORDER BY l_linenumber
           "#,
   ```


-- 
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: github-unsubscr...@datafusion.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org

Reply via email to