-
Notifications
You must be signed in to change notification settings - Fork 217
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
Restoration bench: increase heap and only bench {0%,0.2%} x {seq,rnd} #2688
Conversation
It seems the final listTransactions benchmark at the end surpasses the 8GB of memory. A quick workaround could be to just bump the limit. We can investigate the memory usage of listing transactions separately if we deem it worthfile.
lib/shelley/bench/Restore.hs
Outdated
@@ -313,7 +313,8 @@ cardanoRestoreBench tr c socketFile = do | |||
, benchRestoreSeqWithOwnership (Proxy @4) | |||
|
|||
, benchRestoreRndWithOwnership (Proxy @0) | |||
, benchRestoreRndWithOwnership (Proxy @1) | |||
-- benchRestoreRndWithOwnership (Proxy @1) | |||
-- Removed to prevent timeout in CI. |
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.
Keeping it commented to prevent someone from thinking the absence of it is a mistake.
And I kept 0.4% rnd, because I recall it being similar to some exchanges (but I don't really know).
Looking at the current progress, removing 0.1%-rnd will most certainly not be enough. Before removing anything, it started 0.4%-rnd 30 minutes before the timeout. So even just removing 0.4%-rnd will be tight. I suspect we might need to remove both 0.4%-seq and 0.4%-rnd… |
From 0.4%-seq src 😳 |
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.
Thanks for fixing it - looks good to me.
|
||
, benchRestoreRndWithOwnership (Proxy @0) | ||
, benchRestoreRndWithOwnership (Proxy @1) | ||
, benchRestoreRndWithOwnership (Proxy @4) | ||
, benchRestoreRndWithOwnership (Proxy @2) |
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.
It's a little bit confusing to have the address discovery ratio as a unitless type parameter, but now I see this means 0% and 0.2%.
Probably just one or two percentage benchmarks is enough anyway.
Next thing we should do is improve the performance so we can increase these numbers. 😄.
And also to benchmark the sequential address pool code in isolation.
bors r+ |
2688: Restoration bench: increase heap and remove 0.1%-rnd r=rvl a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Johannes Lund <[email protected]>
Build failed: |
bors r+ |
2688: Restoration bench: increase heap and remove 0.1%-rnd r=Anviking a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Johannes Lund <[email protected]>
Build failed:
But there's no more details 🤔 |
bors r+ |
2688: Restoration bench: increase heap and only bench {0%,0.2%} x {seq,rnd} r=Anviking a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Johannes Lund <[email protected]>
Build failed: |
bors r+ |
2686: cabal: Add version constraint on random r=rvl a=rvl # Issue Number ADP-961 # Overview In `random>=1.2.0`, `StdGen` no longer has a `Read` instance. - Add version constraint so Cabal uses previous version. - Clean up `cabal.project`, with the assumption of Cabal >= 3.4.0.0. - Add a nightly CI step to run the cabal build. - Use cabal-cache to make the nightly build faster. - Make sure every Haskell tool built for the nix-shell has a materialization, for faster evaluation without warnings. # Comments Resolves #2685. We may fix the usage of `instance Read StdGen` later. 2688: Restoration bench: increase heap and only bench {0%,0.2%} x {seq,rnd} r=rvl a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Rodney Lorrimar <[email protected]> Co-authored-by: Johannes Lund <[email protected]>
Build failed (retrying...):
2 of the pools have stake, but the third one doesn't. So a problem in the test setup. |
2688: Restoration bench: increase heap and only bench {0%,0.2%} x {seq,rnd} r=rvl a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Johannes Lund <[email protected]>
Build failed:
|
bors r+ |
2688: Restoration bench: increase heap and only bench {0%,0.2%} x {seq,rnd} r=Anviking a=Anviking # Issue Number ADP-846 # Overview Let the nightly benchmark finish successfully within the alotted time. - [x] Increase max heap from 8GB to 16GB - [x] Replace 0.1% and 0.4% rnd and seq benchmarks with 0.2% to prevent timeout. # Comments [![Build status](https://badge.buildkite.com/59ea9363b8526e867005ca8839db47715bc5f661f36e490143.svg?branch=anviking%2FADP-846%2Fmore-heap)](https://buildkite.com/input-output-hk/cardano-wallet-nightly/builds?branch=anviking%2FADP-846%2Fmore-heap) Co-authored-by: Johannes Lund <[email protected]>
Build failed:
|
bors r+ |
Build succeeded: |
Issue Number
ADP-846
Overview
Let the nightly benchmark finish successfully within the alotted time.
Comments