Refactor the bandwidth logging to be less magic #1670
Merged
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
ordian
reviewed
Jul 21, 2020
src/bandwidth.rs
Outdated
Comment on lines
135
to
136
| download: Atomic<u64>, | ||
| upload: Atomic<u64>, |
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
romanb
approved these changes
Jul 22, 2020
I'm not sure I understand your question, but this change is mostly in preparation for |
Co-authored-by: Roman Borschel <romanb@users.noreply.github.com>
hrxi
added a commit
to hrxi/rust-libp2p
that referenced
this pull request
May 11, 2022
Replace `atomic::Atomic<u64>` by `std::sync::atomic:AtomicU64`. The original motivation of using `atomic::Atomic<u64>` instead of `std`'s version in libp2p#1670 was the following: > I used the atomic crate and an Atomic<u64> because the AtomicU64 type > isn't available on all architectures. The atomic crate automatically > falls back to using a Mutex on platforms that don't support AtomicU64. This argumentation is moot because the crate directly depends on `libp2p-core` which also uses `AtomicU64`.
thomaseizinger
pushed a commit
that referenced
this pull request
May 12, 2022
Replace `atomic::Atomic<u64>` by `std::sync::atomic:AtomicU64`. The original motivation of using `atomic::Atomic<u64>` instead of `std`'s version in #1670 was the following: > I used the atomic crate and an Atomic<u64> because the AtomicU64 type > isn't available on all architectures. The atomic crate automatically > falls back to using a Mutex on platforms that don't support AtomicU64. This argumentation is moot because the crate directly depends on `libp2p-core` which also uses `AtomicU64`.
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.
At the moment, the bandwidth logging system in libp2p automatically tries to perform a 5-seconds average of the number of bytes that are transferred through its sockets.
This PR completely reworks that system and instead makes the bandwidth logging system only count and report the number of bytes. The user can later grab this number of bytes and perform their own 5-seconds-average if they need to.
This reason for this change is that users wouldn't be able to plug the
BandwidthLoggingstruct on a QUIC transport. In order to make it possible for users to measure the bandwidth of QUIC connections, we would have to duplicate the averaging-algorithm of theBandwidthLogginginside of the QUIC transport itself.I think it is an overall better approach to simply measure the number of bytes. It removes a
Mutex, and is generally less magic than performing our own calculation.I used the
atomiccrate and anAtomic<u64>because theAtomicU64type isn't available on all architectures. Theatomiccrate automatically falls back to using aMutexon platforms that don't supportAtomicU64.