Fix IOStats for Nimble #10216
Closed
Fix IOStats for Nimble #10216
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
✅ Deploy Preview for meta-velox canceled.
|
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
4247f01 to
440d444
Compare
helfman
added a commit
to helfman/velox
that referenced
this pull request
Jun 17, 2024
Summary: Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta Differential Revision: D58559606
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
helfman
added a commit
to helfman/velox
that referenced
this pull request
Jun 17, 2024
Summary: Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta Differential Revision: D58559606
440d444 to
d933c52
Compare
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
helfman
added a commit
to helfman/velox
that referenced
this pull request
Jun 17, 2024
Summary: Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
d933c52 to
b414aeb
Compare
helfman
added a commit
to helfman/nimble
that referenced
this pull request
Jun 20, 2024
Summary: X-link: facebookincubator/velox#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
b414aeb to
82f9f4e
Compare
helfman
added a commit
to helfman/velox
that referenced
this pull request
Jun 20, 2024
Summary: Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
Yuhta
approved these changes
Jun 21, 2024
helfman
added a commit
to helfman/nimble
that referenced
this pull request
Jun 22, 2024
Summary: Pull Request resolved: facebookincubator#65 X-link: facebookincubator/velox#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
82f9f4e to
ac9fb99
Compare
helfman
added a commit
to helfman/velox
that referenced
this pull request
Jun 22, 2024
Summary: X-link: facebookincubator/nimble#65 Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
helfman
added a commit
to helfman/nimble
that referenced
this pull request
Jun 22, 2024
Summary: Pull Request resolved: facebookincubator#65 X-link: facebookincubator/velox#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
Summary: X-link: facebookincubator/nimble#65 Pull Request resolved: facebookincubator#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606
|
This pull request was exported from Phabricator. Differential Revision: D58559606 |
ac9fb99 to
9e126c1
Compare
|
This pull request has been merged in 1ae6224. |
facebook-github-bot
pushed a commit
to facebookincubator/nimble
that referenced
this pull request
Jun 26, 2024
Summary: Pull Request resolved: #65 X-link: facebookincubator/velox#10216 IOStats are being calculated in different layers of the IO stacks. Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble. Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF). But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it. Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats. Reviewed By: Yuhta, sdruzkin Differential Revision: D58559606 fbshipit-source-id: 7a13710e5273bd07f19106564c86cce88902da38
|
Conbench analyzed the 1 benchmark run on commit There were no benchmark performance regressions. 🎉 The full Conbench report has more details. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
IOStats are being calculated in different layers of the IO stacks.
Since Nimble and DWRF don't share parts of the stack, some IOStats calculation were not affecting Nimble.
Probably the right thing to do is to move all IOStats calculations to the bottom layers (WSFile, cache and SSD reads), where IO is actually performed (and these layers are shared beteen Nimble nad DWRF).
But it seems like that for this change, we need a design, clarifying what we actually want to track and how to track it.
Since we don't have the cycles to create this design right now, I opted for a simple solution, where I create a simple layer on the Nimble side, which will calculate these stats.
Reviewed By: Yuhta
Differential Revision: D58559606