-
Notifications
You must be signed in to change notification settings - Fork 275
chore(docs): update benchmark results and add throughput tables #2905
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
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CPU looks ok, just one thing on the multi bit PBS seems we gained a lot for the CPU is that expected ?
tfhe/docs/getting-started/benchmarks/cpu/cpu-programmable-bootstrapping.md
Show resolved
Hide resolved
1dd7fc3
to
0955fea
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good for CPU, approving here to trigger CI,
I let @agnesLeroy and @pgardratzama check the latency for their devices :)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me, only the GPU division latency needs an update still 🙏
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Wait but I only see latency tables, we should also be adding the throughput tables for integer ops @soonum
throughput has been delayed until we have everything for all devices, my understanding is we were missing some |
0955fea
to
5b7d35d
Compare
tfhe/docs/.gitbook/assets/hpu-integer-benchmark-hpux1-tuniform-2m128-plaintext.svg
Show resolved
Hide resolved
6832f86
to
ed03dc9
Compare
tfhe/docs/.gitbook/assets/cpu-gpu-hpu-integer-benchmark-fheuint64-tuniform-2m128-ciphertext.svg
Show resolved
Hide resolved
ed03dc9
to
54cb2b3
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@soonum for the GPU benchmarks I don't see scalar division in the tables, I'm sure we run it though 🤔
will need a rebase after the value Agnès mentions is fixed @soonum :) |
54cb2b3
to
8264e56
Compare
This also removes tables in PBS benchmarks for failure probability of 2**-40.
8264e56
to
de016e2
Compare
@IceTDrinker I think we're good to merge? |
Yes |
This also removes tables in PBS benchmarks for failure probability of 2**-40.