Scalability and Perf Board issueshttps://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues2020-05-20T05:50:14Zhttps://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/42FlashFlow2020-05-20T05:50:14ZMike PerryFlashFlowhttps://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/41Conflux (unreliable version)2022-01-24T16:51:22ZMike PerryConflux (unreliable version)Conflux without a reconnection window will reduce network reliability, but improve throughput.
It may increase latency due to circuit setup, if circuits can't be pre-built.
It may also provide traffic analysis resistance benefits.Conflux without a reconnection window will reduce network reliability, but improve throughput.
It may increase latency due to circuit setup, if circuits can't be pre-built.
It may also provide traffic analysis resistance benefits.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/29Adaptive Stream Timeout2020-02-07T18:42:29ZMike PerryAdaptive Stream TimeoutSimilar to DNS tests, we could deploy an adaptive stream timeout using CBT style math.Similar to DNS tests, we could deploy an adaptive stream timeout using CBT style math.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/17Consensus param for Fast cutoff2020-07-09T20:14:26ZMike PerryConsensus param for Fast cutoffThe Fast cutoff is currently hard-coded. If we made a consensus param for this value, we could reduce consensus size and also improve long-tail performance by eliminating slow relays.The Fast cutoff is currently hard-coded. If we made a consensus param for this value, we could reduce consensus size and also improve long-tail performance by eliminating slow relays.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/15Improve relay bandwidth testing/self-reporting2020-02-07T18:18:38ZMike PerryImprove relay bandwidth testing/self-reportingWe could improve relay onboarding with a faster/more accurate/more trusted measurement step.We could improve relay onboarding with a faster/more accurate/more trusted measurement step.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/14Tweak sbws scaling curves2020-02-07T18:18:10ZMike PerryTweak sbws scaling curvesAs an alternative, or perhaps in addition to, PID feedback, we could tweak the weights that sbws applies by imposing an additional curve on top of the results.
I am not a huge fan of this approach, but I am adding it for completeness.As an alternative, or perhaps in addition to, PID feedback, we could tweak the weights that sbws applies by imposing an additional curve on top of the results.
I am not a huge fan of this approach, but I am adding it for completeness.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/13Instrumentation/tracing framework2020-02-07T17:07:34ZMike PerryInstrumentation/tracing frameworkDgoulet believes we can more easily find long-tail perf bugs with a tracing framework for Tor, so we can get high-resolution information about critical paths in relay use without impacting performance during the instrumentation of the To...Dgoulet believes we can more easily find long-tail perf bugs with a tracing framework for Tor, so we can get high-resolution information about critical paths in relay use without impacting performance during the instrumentation of the Tor process.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/12PID feedback for bw auths2020-02-07T18:17:32ZMike PerryPID feedback for bw authsThe Torflow PID feedback system will result in a provably optimimum load balancing equilibrium for uniform average stream capacity across the network.
The problem is if we only use bandwidth for this feedback, relays get overloaded in o...The Torflow PID feedback system will result in a provably optimimum load balancing equilibrium for uniform average stream capacity across the network.
The problem is if we only use bandwidth for this feedback, relays get overloaded in other ways. So we need to stop feedback if nodes announce that they are overloaded via memory, sockets, or CPU (which is another ticket).https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/11DNS test predicted circuits2020-02-07T18:15:35ZMike PerryDNS test predicted circuitsArthur and Dennis have detected that some long tail of exits/relays are really bad for unknown reasons. Sometimes this is just DNS config, but there are other things beyond that.
We could DNS test our circuits and toss out the slowest 5...Arthur and Dennis have detected that some long tail of exits/relays are really bad for unknown reasons. Sometimes this is just DNS config, but there are other things beyond that.
We could DNS test our circuits and toss out the slowest 5%, via a mechanism similar to CBT math. We don't really need to toss out a lot of circuits via this mechanism, as something like 1% or less of circuits have this problem. But the ones that do have it delay page load by 6 seconds or more.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/10Add relay descriptor field to indicate shared link2020-02-07T18:16:56ZMike PerryAdd relay descriptor field to indicate shared linkWhen Rob did his flashflow experiment, relays that shared a machine or link with other relays got overloaded.
If we could encourage relay operators to set a flag for their multi-instance deployments, or better, auto-detect/guess and the...When Rob did his flashflow experiment, relays that shared a machine or link with other relays got overloaded.
If we could encourage relay operators to set a flag for their multi-instance deployments, or better, auto-detect/guess and then ask for confirmation, we could avoid overloading shared-link relays via either flashflow or sbws.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/9Signals for relay CPU, memory, socket overload2020-02-07T18:13:47ZMike PerrySignals for relay CPU, memory, socket overloadWe need low resolution backoff signals for bw auth PID feedback, so that the bw auths can stop providing more weight to relays that are maxed out in ways other than bandwidth.
Bonus points if can be used by the congestion control resear...We need low resolution backoff signals for bw auth PID feedback, so that the bw auths can stop providing more weight to relays that are maxed out in ways other than bandwidth.
Bonus points if can be used by the congestion control research idea.https://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/3Walking Onions2020-02-10T22:03:25ZMike PerryWalking Onionshttps://gitlab.torproject.org/mikeperry/scalability-and-perf-board/-/issues/1Congestion Control2020-07-09T20:14:22ZMike PerryCongestion Control