Repository navigation
Conversation
…#7) Measures, at n = 1,000 and 10,000 members, both profiles: per-pull handler CPU with the REAL default membership (realm-signed endorsement verified every pull through the injectable lookup), rotation reply bytes, the link-layer costs §6.3 prices a sealed pull at (signed RESULT, KEM decapsulation, UCAN authorize), and publish/open per event. Measured (OTP 28, host00): the handler is ~0.25 ms (pq_pure) and ~0.47 ms (pq_hybrid) per pull; the link layer dominates at ~0.9 + 0.17 + 2.5 ms (pure) and ~6.1 + 0.17 + 3.6 ms (hybrid). Full rotation: ~4.1 s / ~10.5 s at 1k and ~44 s / ~111 s at 10k, against §6.3's 1.3 / 5.5 s and 13 / 55 s. The signature and KEM terms match the estimate; the UCAN authorize is the delta, costing 2.5-4.2 ms where §6.3 assumed ~0.31 ms. The register's numbers should cite these.
This branch has not been deployed
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The first measured numbers for the register's §13 #7: rotation CPU and reply bytes at n = 1,000 and 10,000 members, both profiles, with the real default membership path (realm-signed endorsement verified per pull) and the link-layer costs §6.3 prices a sealed pull at.
Measured (OTP 28, host00, one core):
The signature and KEM terms match the estimate; the UCAN authorize is the delta: 2.5-4.2 ms where §6.3 assumed ~0.31 ms for the CALL verify. Corrected, a 10,000-member EU rotation is ~111 s of distributor CPU (12% of one core at the 15-minute bound), not 55 s.
Known exclusions, stated in the bench: the DHT lookup I/O per pull (latency, station-bounded by D28 budgets, not distributor CPU), wire encoding (term_to_binary proxy for reply bytes), and the jittered re-pull's spread (totals are aggregates, not bursts).
Run:
scripts/bench-group-distributor.sh(env N1/N2), afterrebar3 as test compile.