- Notifications
You must be signed in to change notification settings - Fork 1.8k
[TEST] Improve combine_hashes #19391
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
| run benchmarks |
| run benchmark tpch |
| 🤖 |
| run benchmark tpcds |
| 🤖: Benchmark completed Details
|
| 🤖 |
| 🤖: Benchmark completed Details
|
| 🤖 |
| 🤖: Benchmark completed Details
|
| run benchmarks |
| 🤖 |
| run benchmark tpch tpcds |
| 🤖: Benchmark completed Details
|
| 🤖 |
| 🤖: Benchmark completed Details
|
| 🤖 |
| 🤖: Benchmark completed Details
|
| run benchmark with_hashes |
| 🤖 |
| Benchmark script failed with exit code 101. Last 10 lines of output: Click to expand |
| run benchmark tpch tpcds |
| 🤖 |
| 🤖: Benchmark completed Details
|
| 🤖 |
| 🤖: Benchmark completed Details
|
| This does not show much difference (the idea is that the current |
| run benchmark with_hashes |
| 🤖 |
| (I am trying to run the hash micro benchmark to see if we can see any difference) |
| 🤖: Benchmark completed Details
|
| If anything, it should be a bit slower (because of extra instructions). I wrote a test that shows this method to be better when the resulting hash output is of low quality (only bits on the right side), but normal "random" bits seem to work just fine for the usecase of partitioning. It would be probably even better to avoid using a hash combine method and directly use the hasher on multiple columns 🤔 |
Which issue does this PR close?
Rationale for this change
What changes are included in this PR?
Are these changes tested?
Are there any user-facing changes?