]> 2026-09-08T17:55:24+09:00 Lesson:scale and flexibility in distribution of hot content a83ac58f 3000 ko Most directly applicable to high-fanout object distribution inside a managed private cloud. The official abstract does not establish equivalent behavior on open or adversarial networks. Owl combines ephemeral peer-to-peer distribution trees in a decentralized data plane with centralized tracker services. Trackers maintain peer/cache/download metadata, choose fetch sources and retries, control caching/eviction, and expose a configurable policy interface; peers remain simple state machines. S3ResearchAgent medium Venue: OSDI. Year: 2022. Meta needed one distribution service for millions of client processes and heterogeneous workloads. Verification: official USENIX OSDI 2022 paper page/abstract. 2026-07-16T14:50:12Z 2461238.1181944 Scale and Flexibility in Distribution of Hot Content. OSDI 2022. Centralizing control decisions need not prevent data-plane scalability when bulk transfer remains peer-to-peer. Scale and Flexibility in Distribution of Hot Content USENIX reports more than 800 PB distributed per day, 2–3× faster downloads than BitTorrent and Meta’s prior decentralized static tree, and production support for 106 use cases using 55 policies. How can a private cloud distribute large, hot objects at very high fanout while supporting diverse per-use-case policies? lab Centralizing control decisions need not prevent data-plane scalability when bulk transfer remains peer-to-peer. Draft 2026-07-18T14:59:02Z 2461240.1243287 2026-07-18T14:59:02Z 2461240.1243287 scale and flexibility in distribution of hot content a83ac58f 3000 ev_0854bdeb3cdb4f0e 2026-07-16T14:50:19Z 2461238.1182755 S3ResearchAgent Scale and Flexibility in Distribution of Hot Content. OSDI 2022. paper Bibliographic paper record. not_recorded scale and flexibility in distribution of hot content a83ac58f# a73ac98efe6b156b29c678fd34d3c3e9 3000 verified-content-v1-0001 2026-07-16T18:34:01Z 2461238.2736227 S3ResearchAgent Jason Flinn et al., "Owl: Scale and Flexibility in Distribution of Hot Content," OSDI 2022. paper Verification scope: official USENIX OSDI 2022 paper page/abstract. Research question: How can a private cloud distribute large, hot objects at very high fanout while supporting diverse per-use-case policies? Context: Meta needed one distribution service for millions of client processes and heterogeneous workloads. Method: Owl combines ephemeral peer-to-peer distribution trees in a decentralized data plane with centralized tracker services. Trackers maintain peer/cache/download metadata, choose fetch sources and retries, control caching/eviction, and expose a configurable policy interface; peers remain simple state machines. Evaluation/results: USENIX reports more than 800 PB distributed per day, 2–3× faster downloads than BitTorrent and Meta’s prior decentralized static tree, and production support for 106 use cases using 55 policies. Interpretation: Centralizing control decisions need not prevent data-plane scalability when bulk transfer remains peer-to-peer. Applicability/limits: Most directly applicable to high-fanout object distribution inside a managed private cloud. The official abstract does not establish equivalent behavior on open or adversarial networks. not_recorded scale and flexibility in distribution of hot content a83ac58f# d65ac35f3ecabe25022632af23ca9e1e 3000 canonical-paper-v2-a83ac58f 2026-07-18T05:36:30Z 2461239.7336806 S3ResearchAgent Jason Flinn et al., "Owl: Scale and Flexibility in Distribution of Hot Content," OSDI 2022. paper Verification scope: official USENIX OSDI 2022 paper page/abstract; confidence=medium. Research question: How can a private cloud distribute large, hot objects at very high fanout while supporting diverse per-use-case policies? Context: Meta needed one distribution service for millions of client processes and heterogeneous workloads. Method: Owl combines ephemeral peer-to-peer distribution trees in a decentralized data plane with centralized tracker services. Trackers maintain peer/cache/download metadata, choose fetch sources and retries, control caching/eviction, and expose a configurable policy interface; peers remain simple state machines. Evaluation/results: USENIX reports more than 800 PB distributed per day, 2–3× faster downloads than BitTorrent and Meta’s prior decentralized static tree, and production support for 106 use cases using 55 policies. Interpretation: Centralizing control decisions need not prevent data-plane scalability when bulk transfer remains peer-to-peer. Applicability/limits: Most directly applicable to high-fanout object distribution inside a managed private cloud. The official abstract does not establish equivalent behavior on open or adversarial networks. official_abstract scale and flexibility in distribution of hot content a83ac58f# 09cdf071bb881444871d081336a1f248 3000 2026-07-18T14:59:02Z 2461240.1243287 S3ResearchAgent verify_dddd8c54a48f0cbb52b0 partial_source context 보존 원문 객체를 확보했으나 이 일괄 검증에서는 claim-bearing 범위를 재판정하지 않아 결론을 보류함. inconclusive 1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 보존 파일 objects/sha256/1c/1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 ev_0854bdeb3cdb4f0e 7e4ade75e01d3decccb118c9f6ee6cb9cdb7a452c3e211f4309719cd041bfc58 scale and flexibility in distribution of hot content a83ac58f# 2c8a18620912460cbfc366cbd39fcbce 3000 2026-07-18T14:59:02Z 2461240.1243287 S3ResearchAgent verify_caf7b25e2e28b7b2859a partial_source context 보존 원문 객체를 확보했으나 이 일괄 검증에서는 claim-bearing 범위를 재판정하지 않아 결론을 보류함. inconclusive 1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 보존 파일 objects/sha256/1c/1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 verified-content-v1-0001 7bd37421539f6e26bfd7eb455f50629a63817915a84563eb66aa9d15b0b3dd0e scale and flexibility in distribution of hot content a83ac58f# 5e257707b0645392e06de825ec2bc19b 3000 2026-07-18T14:59:02Z 2461240.1243287 S3ResearchAgent verify_33735330ceaa4151bdf1 partial_source context 보존 원문 객체를 확보했으나 이 일괄 검증에서는 claim-bearing 범위를 재판정하지 않아 결론을 보류함. inconclusive 1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 보존 파일 objects/sha256/1c/1cbe42f4f0a0a1dbbb509f57708823da5b87f6a908261467c3f346e6df757432 canonical-paper-v2-a83ac58f cf8e0d16413745df863f7feff1d91a97082a2c3b67fd01f88d9d281ad4ba049f scale and flexibility in distribution of hot content a83ac58f# a7d9fe40165a7ee8164745381a2cf704