Daan MeijerandClaude Opus 5.5 69f1cc63b2 Rebuild: coordinator:run --shard=i/n splits prefetching over processes ...

The coordinator's event loop was 16.5% busy at ~10 MB/s of compressed
XOVER (socket reads, buffers, GC, spool writes; nothing stands out), so
one process tops out around 60 MB/s: less than 50 connections bring
(~1M articles/s, ~70-86 MB/s). A shard prefetches the ranges with
id % n = i - 1 with its own NNTP_CONNECTIONS; only shard 1 plans, merges
and settles. Two shards of 5 connections fetched what one of 10 did.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-09 03:36:24 +02:00

..

batch_store.ts

Rebuild: the fragment merger

2026-10-09 02:19:53 +02:00

chunk_partitions.ts

Rebuild: coordinator (planner, NNTP pool, spool, prefetcher, leases, rule sets), node API and the edge:node worker

2026-10-09 00:31:06 +02:00

fragment_merger.ts

Rebuild: index lookups for the merger's neighbours and the prefetcher's ranges

2026-10-09 02:59:30 +02:00

leases.ts

Rebuild: a lease never waits for prefetching; one-shot hashes

2026-10-09 03:08:19 +02:00

nntp_pool.ts

Rebuild: the coordinator stops on SIGTERM

2026-10-09 03:20:11 +02:00

nzb_service.ts

Rebuild: NZBs from the chunk store, and a catalog search

2026-10-09 02:33:27 +02:00

planner.ts

Rebuild: merger rounds with one neighbour query, a cursor and a time budget; planning skips failing groups

2026-10-09 02:25:45 +02:00

prefetcher.ts

Rebuild: coordinator:run --shard=i/n splits prefetching over processes

2026-10-09 03:36:24 +02:00

rule_sets.ts

Rebuild: first end-to-end runs; prefetch per chunk, executable rule, wire fixes

2026-10-09 01:07:11 +02:00

settler.ts

Rebuild: settling quiet releases

2026-10-09 02:33:47 +02:00

spool.ts

Rebuild: coordinator (planner, NNTP pool, spool, prefetcher, leases, rule sets), node API and the edge:node worker

2026-10-09 00:31:06 +02:00