-
Notifications
You must be signed in to change notification settings - Fork 112
Conversation
7f0b951
to
33d1de2
Compare
due to issues with test suite rebased on to bug fixed branch and this is PR'd against that for now. |
33d1de2
to
9383abb
Compare
make both performance benchmarks write to a tmp dir and put in the .gitignore
Limits connection bandwidth in real world benchmarks so that blocks are delayed if single peer is overused fix #40
9383abb
to
fe0a253
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.
LGTM but I'd consider running gx update QmYxivS34F2M2n44WQQnRHGAKS8aoRUxwGpi9wk4Cdn4Jf
to update go-libp2p for the rate limiter stuff (so we don't have to do it later).
testnet/virtual.go
Outdated
|
||
type rateLimiter interface { | ||
Limit(dataSize int) time.Duration | ||
} |
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.
Yeah, that type should be exported: libp2p/go-libp2p#507. If you want, you can update go-libp2p to 6.0.30 to pull this in.
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.
Oh, nice :)
Updated Libp2p and used it's newly exposed RateLimiter public interface
…d-tests Feat/bandwidth limited tests This commit was moved from ipfs/go-bitswap@18c401d
Goals
Existing benchmarks do not take into account bandwidth limitations on a peer. IOW, assuming all peers have all blocks for a benchmark, the perfect result can be obtained by just continuing to ask the single fastest peer for blocks.
Implementation