v0.1.0

Changelog

Other

• 6d04bb9: Add M0: data model, .meguri format, inspect, and the full pipeline (#1) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 8472c6e: Add M3 politeness, DNS, and robots (#4) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 7cd0e13: Add URL canonicalization and registrable-domain keying (tamnd 1218621+tamnd@users.noreply.github.com)
• eabcebb: Add a bloom-model arm to the baseline comparison (#45) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• fd2ed18: Add a projection and predicate-pushdown reader for .meguri (#16) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 75b9d8f: Add calendar, faceted, and session-id trap heuristics (#19) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 2023299: Add content fingerprinting and near-dup detection (tamnd 1218621+tamnd@users.noreply.github.com)
• 0f436df: Add freshness offline replay with the four scheduling baselines (#20) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• f55742d: Add resident due-time schedule index (timing wheel) (#31) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 76a06a1: Add stats and map inspection commands (#46) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• d237af9: Add the FSST per-ref string codec and arena (#27) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 6fa14be: Add the M1 single-partition frontier scheduler (#2) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• a36d49a: Add the M10 benchmark assembly and the 100B projection (#12) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 4e325a9: Add the M2 corpus gate and benchmarks on real data (tamnd 1218621+tamnd@users.noreply.github.com)
• 9e6de97: Add the M4 freshness rescheduler (#5) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• e8631de: Add the M5 prioritizer: OPIC, import blend, STAR budget (#6) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 6d93bba: Add the M6 durable storage engine (#7) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• ce9c583: Add the M7 codec cascade, string blob, manifest, and file ops (#9) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• bff65e6: Add the M8 distribution layer: jump-hash map, router, and file-shipping rebalance (#10) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• bd8c03b: Add the M9 replication and failover layer (#11) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 858f140: Add the bounded-frontier anti-starvation tier (#23) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 897ac89: Add the columnar delta+FSST discovery batch body (#29) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• a3333ef: Add the durable partition lifecycle and serve command (#35) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• a4de7ed: Add the elasticity control loop over the existing resize mechanisms (#26) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• f5e86d8: Add the outcome state machine: retry, tombstone, redirect edges (#14) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 7bd204b: Add the pack and compact file commands (#48) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 8e27863: Add the ribbon static seen-set filter (#18) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 8f15e62: Add the schedule index region to the format (#17) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• c72a758: Add the scheduler-vs-politeness throughput analysis (#21) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• c9afb38: Add the seen-set baseline against a naive exact-key frontier (#22) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• e4856de: Add the ship-then-delete rebalance handoff (#50) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 1ab3008: Add the staged single-writer engine and the seed adapters (#13) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 7000555: Add the string-region dictionary and robots blob packing (#28) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 387782c: Add the tiered seen-set with a DRUM batch path (tamnd 1218621+tamnd@users.noreply.github.com)
• 894c53d: Add trap defense: depth and budget admission with soft-404 detection (tamnd 1218621+tamnd@users.noreply.github.com)
• e8f6e83: Bind the ami fetcher to the engine through an outcome adapter (#40) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• aa13a92: Bind the elasticity loop to backlog reported on heartbeats (#38) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 01431af: Coalesce cross-partition discoveries across outcomes (#30) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 28bb5f2: Gate checkpoint reclamation on the corpus (#43) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 001843f: Gate the rank import against a real PageRank computation (#42) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 3a6b1b1: Initial commit (tamnd 1218621+tamnd@users.noreply.github.com)
• 2de1538: Measure rebalance shipped bytes against the bandwidth wall (#39) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 5efe6fa: Measure routed discovery intake throughput on the corpus (#41) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 925cbc6: Pin release actions and cosign to fixed versions (#8) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 3a8c241: Pin the benchmark corpus as a host-key range (#34) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• d1c05f9: Pin the corpus identity as a checked guard (#44) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 5ed5d46: Pin the host-auto-create and schedule-membership invariants (#33) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 38a0909: Rewrite docs for the v0.1.0 release (#51) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 56d2283: Route tsumugi signal imports per owning partition (#25) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 97a4dbc: Serialize the seen-set filter into the .meguri file (#15) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 5667b29: Shard the ready-host bank for work-stealing dispatch (#32) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 52ff36b: Share one resolver cache across a machine's partitions (#24) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 27ab282: Split columns into pages with a per-page skip list (#37) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• a568ec3: Stamp the source host on out-links so cross-partition in-degree is correct (#36) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• e1f1df0: Wire sitemap parse output into frontier discoveries (#49) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 5583afd: Wire the durable schedule wheel into checkpoint and add the schedule command (#47) (Tam Nguyen Duc 1218621+tamnd@users.noreply.github.com)
• 2445d7d: Wire the seen-set into frontier discovery and reporting (tamnd 1218621+tamnd@users.noreply.github.com)