Changes between Version 32 and Version 33 of org/roadmaps/CoreTor/PerformanceExperiments


Ignore:
Timestamp:
May 9, 2019, 8:08:13 PM (3 months ago)
Author:
mikeperry
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • org/roadmaps/CoreTor/PerformanceExperiments

    v32 v33  
    6363  * If the TIMEOUT_RATE field or the manually counted circuit timeout rate exceeds 1.0-CUTOFF_QUANTILE by more than 0.1, the experiment should be stopped and we should investigate and debug the circuit build timeout code.
    6464  * The failure rate of torperf and onionperf should be closely monitored, to ensure that onion services do not have unexpected amounts of additional failure during this experiment. If failure rates increase, we should abort.
    65   * If Torperf reduces the set of exit nodes or rendezvous points used, or alters their frequency significantly from consensus weights, we should abort.
     65  * If Torperf uses exit nodes and rendezvous points out of proportion to their consensus weights after this change, we should abort. [https://github.com/mikeperry-tor/vanguards/blob/master/src/vanguards/rendguard.py The vanguards rendguard component] has code to monitor rend point use already; it can be adapted for exits as well.
    6666* **User Impact/What to Tell Users:**
    6767  * So long as the abort criteria are not met, the user impact should be minimal for small changes to this parameter: latency should just improve. For very low values, the geolocation concern increases, but we should be able to rule those out through the abort criteria.