]> 2026-09-08T15:37:44+09:00 Lesson:flexmem adaptive page profiling and migration for tiered memory 43991e2a 3000 ko Linux heterogeneous/tiered-memory systems. Limits: Tuning and gains depend on hardware counters, memory topology, migration costs, and workload phase length. FlexMem combines performance counters with hint faults, dynamically adjusts demotion volume, and tracks warm-page ranges. S3ResearchAgent high Venue: USENIX ATC. Year: 2024. Fixed sampling and promotion/demotion thresholds either miss hot pages or spend too much on profiling and movement. Verification: full_text; confidence=high. 2026-07-16T15:04:44Z 2461238.128287 FlexMem: Adaptive Page Profiling and Migration for Tiered Memory. USENIX ATC 2024. Profiling fidelity and migration aggressiveness should be controlled together from workload feedback. FlexMem: Adaptive Page Profiling and Migration for Tiered Memory workloads=common memory-intensive benchmarks; baselines=Tiering-0.8; TPP; MEMTIS; metrics=application performance; profiling overhead; migration efficiency; results=32% average over Tiering-0.8; 23% over TPP; 27% over MEMTIS How can tiered memory adapt profiling and migration to workloads whose hot sets and phase behavior differ? lab Adapt both measurement cost and actuation strength, not just the placement threshold. Draft 2026-07-18T14:58:33Z 2461240.1239931 2026-07-18T14:58:33Z 2461240.1239931 flexmem adaptive page profiling and migration for tiered memory 43991e2a 3000 ev_f61b5cc359014df0 2026-07-16T15:04:45Z 2461238.1282986 S3ResearchAgent FlexMem: Adaptive Page Profiling and Migration for Tiered Memory. USENIX ATC 2024. paper Bibliographic paper record. not_recorded flexmem adaptive page profiling and migration for tiered memory 43991e2a# 1770b7b3f58a4980356ad2e415b513f5 3000 verified-content-v1-0083 2026-07-16T18:47:24Z 2461238.2829167 S3ResearchAgent Dong Xu et al., "FlexMem: Adaptive Page Profiling and Migration for Tiered Memory", USENIX ATC 2024. paper Verification: full_text; confidence=high. Canonical title: FlexMem: Adaptive Page Profiling and Migration for Tiered Memory Question: How can tiered memory adapt profiling and migration to workloads whose hot sets and phase behavior differ? Context: Fixed sampling and promotion/demotion thresholds either miss hot pages or spend too much on profiling and movement. Method: FlexMem combines performance counters with hint faults, dynamically adjusts demotion volume, and tracks warm-page ranges. Evaluation: workloads=common memory-intensive benchmarks; baselines=Tiering-0.8; TPP; MEMTIS; metrics=application performance; profiling overhead; migration efficiency; results=32% average over Tiering-0.8; 23% over TPP; 27% over MEMTIS Interpretation: Profiling fidelity and migration aggressiveness should be controlled together from workload feedback. Reusable lesson: Adapt both measurement cost and actuation strength, not just the placement threshold. Applicability: Linux heterogeneous/tiered-memory systems. Limits: Tuning and gains depend on hardware counters, memory topology, migration costs, and workload phase length. not_recorded flexmem adaptive page profiling and migration for tiered memory 43991e2a# 5d85b0f61f223677b51cf08523e2f6a0 3000 canonical-paper-v2-43991e2a 2026-07-18T05:22:42Z 2461239.7240972 S3ResearchAgent Dong Xu et al., "FlexMem: Adaptive Page Profiling and Migration for Tiered Memory", USENIX ATC 2024. paper Verification: full_text; confidence=high. Canonical title: FlexMem: Adaptive Page Profiling and Migration for Tiered Memory Question: How can tiered memory adapt profiling and migration to workloads whose hot sets and phase behavior differ? Context: Fixed sampling and promotion/demotion thresholds either miss hot pages or spend too much on profiling and movement. Method: FlexMem combines performance counters with hint faults, dynamically adjusts demotion volume, and tracks warm-page ranges. Evaluation: workloads=common memory-intensive benchmarks; baselines=Tiering-0.8; TPP; MEMTIS; metrics=application performance; profiling overhead; migration efficiency; results=32% average over Tiering-0.8; 23% over TPP; 27% over MEMTIS Interpretation: Profiling fidelity and migration aggressiveness should be controlled together from workload feedback. Reusable lesson: Adapt both measurement cost and actuation strength, not just the placement threshold. Applicability: Linux heterogeneous/tiered-memory systems. Limits: Tuning and gains depend on hardware counters, memory topology, migration costs, and workload phase length. full_text flexmem adaptive page profiling and migration for tiered memory 43991e2a# 1f42a33ff94bb10734aab50ef73b53e2 3000 2026-07-18T14:58:32Z 2461240.1239815 S3ResearchAgent verify_3813130dab6423b22d92 partial_source context 보존 원문 객체를 확보했으나 이 일괄 검증에서는 claim-bearing 범위를 재판정하지 않아 결론을 보류함. inconclusive 6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 보존 파일 objects/sha256/6f/6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 ev_f61b5cc359014df0 40f084fe4b7d05e0f92db08420d517bda9a7c29b953f3d99f026406245ebe3ae flexmem adaptive page profiling and migration for tiered memory 43991e2a# 744e598cf1313da477c0f0165ed2191c 3000 2026-07-18T14:58:33Z 2461240.1239931 S3ResearchAgent verify_0ed3556adb542ac293db partial_source context 보존 원문 객체를 확보했으나 이 일괄 검증에서는 claim-bearing 범위를 재판정하지 않아 결론을 보류함. inconclusive 6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 보존 파일 objects/sha256/6f/6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 verified-content-v1-0083 ee9794538e4f9c0240afc9e6be5b1ac28d3a07f8c3a4c14ccef6286f76a3650e flexmem adaptive page profiling and migration for tiered memory 43991e2a# 71d9c86170a8fd85b79ce252dfc9da47 3000 2026-07-18T14:58:33Z 2461240.1239931 S3ResearchAgent verify_daef997e070e4d8a8de3 official_abstract observation,interpretation,reusable_lesson observation=supported; interpretation=supported; reusable_lesson=supported supports 6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854 R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=6fdd7f62fcaf8233cd821d64d6a998ba54ca4e82016eea76d064c42e708e3854; independently adjudicated claim-bearing primary source Saved USENIX presentation page, official abstract. canonical-paper-v2-43991e2a 2b1b6b8073e3db689e4f2e321cd853757ca396d5aaaa2caf9f75bfbb450abfaf flexmem adaptive page profiling and migration for tiered memory 43991e2a# af7e8e4b1244de7ca5d51f1e85add838