aboutsummaryrefslogtreecommitdiffstats
path: root/src/swarm
Commit message (Collapse)AuthorAgeFilesLines
* no swarm ID in synctv11 days2-24/+9
|
* announce only to those behindtv11 days2-12/+31
|
* measure sync envelopestv11 days2-0/+24
|
* dashboard: swarm-wide epoch shinetv11 days1-8/+31
|
* asked known elsewheretv11 days3-2/+23
|
* predecessors go alongtv11 days1-3/+18
|
* covered neighbours have ittv11 days3-16/+54
|
* new changes per messagetv11 days2-1/+10
|
* neigbours by transporttv11 days1-1/+21
|
* own tree shapetv11 days1-10/+136
|
* drop full mesh stattv11 days1-3/+7
|
* delta from held headstv11 days1-1/+4
|
* offer again ungeneratedtv11 days1-32/+53
|
* stale delta offerstv11 days1-0/+19
|
* blob window growstv11 days1-5/+10
|
* delta known to sendertv11 days1-1/+7
|
* delta propagationtv11 days2-2/+18
|
* one delta at a timetv11 days2-9/+53
|
* time blob messagestv11 days4-4/+46
|
* log why a deltatv11 days1-0/+6
|
* offered deltas lapsetv11 days1-4/+14
|
* throughput from delivered messagetv11 days2-2/+59
|
* hop delaystv11 days4-2/+128
|
* benchmark delays by pathtv11 days2-0/+15
|
* fix: change publisher stops with the runtimetv11 days1-3/+8
|
* benchmark what sync bytes aretv11 days2-3/+30
|
* changes from a peer never go backtv11 days1-0/+4
|
* older delivered messages leave us leveltv11 days1-1/+3
|
* benchmark where twice comes fromtv11 days2-4/+11
|
* resend only what is neededtv11 days1-1/+1
|
* benchmark delivered and twicetv11 days2-0/+30
|
* any message taken as deliveredtv11 days2-13/+23
|
* slow round back offtv11 days2-5/+86
|
* tree messages need no answertv11 days3-6/+40
|
* hello again when heard fromtv11 days2-2/+19
| | | | | | While we wait for a peer's HELLO_REPLY, anything else that comes from it makes our HELLO go again at once, at most once a second, instead of when the backoff says.
* unanswered pull frees the slottv12 days2-1/+23
| | | | | | When the answer to a pull is overdue (the sync reply timeout: 1 s in the simtests, 10 s in production), the pull gives up the slot, and the next waiting neighbour is pulled at once.
* overdue sync asked after from thentv12 days2-4/+4
|
* pull after gracetv12 days2-4/+33
| | | | | | | | Before pulling, a node waits one measured round trip to the announcer (1 s at most) and pulls only what is still missing. If the round trip isn't known yet, it pulls at once, as now. On the clean benchmark links the round trip is a few milliseconds, so those scenarios shouldn't change.
* announcements answered when lackingtv12 days4-10/+34
| | | | | | | A neighbour now answers an announcement only if it lacks some of the announced heads. Heads that merely differ are no longer enough, since they do while writes go on. The check reads the document, so it runs off the message consumer, as the pull check already does.
* unanswered members not reachedtv13 days2-14/+43
|
* links: only news pass ontv13 days1-6/+12
|
* dashboard: show writer treestv13 days1-26/+70
|
* writers from authorstv13 days3-27/+46
|
* reach set for the sessiontv13 days2-16/+10
|
* tree changes had alreadytv13 days1-4/+14
|
* announcement round counterstv13 days2-8/+23
|
* docs after treestv13 days1-5/+6
|
* announce what trees misstv13 days2-35/+115
|
* order members by distancetv13 days2-26/+59
|
* replace eager links by treestv13 days6-137/+35
|