속성으로 검색

이 문서는 속성과 이름이 지정된 값으로 설명된 개체를 찾기 위한 간단한 탐색 인터페이스를 제공합니다. 그 밖에 사용 가능한 검색 인터페이스에는 문서 속성 검색 및 Ask 쿼리 빌더가 있습니다.

결과 목록

• Lesson:overcoming the memory wall with cxl enabled ssds ef2155d6  + (workloads=real-world memory traces; baselines=conventional SSD/memory expansion designs; metrics=access latency; estimated lifetime; results=68–91% of requests under 1 μs; estimated lifetime ≥3.1 years.)
• Lesson:infinigen efficient generative inference of large language models with dynamic kv cache manageme a1ca3228  + (workloads=representative LLMs; KV-offloading inference system; baselines=prior KV cache management methods; metrics=inference speed; accuracy; results=Up to 3.00× speedup with higher accuracy.)
• Lesson:faasmem improving memory efficiency of serverless computing with memory pool architecture b6ab9eec  + (workloads=serverless function workloads; baselines=local-memory serverless deployment; metrics=local memory use; deployment density; p95 latency; results=9.9–79.8% lower average local memory; 108–218% higher density; negligible p95 increase)
• Lesson:not a dpu in name only unleashing rdma capable dpus in multi tenant serverless clouds with nadin 7031318c  + (workloads=serverless ingress/data-plane workloads; baselines=host-centric serverless networking; metrics=requests/s, latency, host CPU cores; results=20.9x RPS; up to 21x lower latency; saves up to 7 CPU cores with 2 DPU cores)
• Lesson:midas minimizing write amplification in log structured systems through adaptive group number and 1cac80b9  + (workloads=simulation; flash SSD proof-of-concept; baselines=state-of-the-art garbage collection; metrics=write amplification; throughput; CPU/memory overhead; results=25% lower WAF; 54% higher throughput.)
• Lesson:rl watchdog a fast and predictable ssd liveness watchdog on storage systems a9a43b61  + (workloads=single SSD; RAID; realistic power-fault injection; baselines=fixed-timeout watchdogs; metrics=data loss; failure-point accuracy; detection predictability; results=96.7% less data loss; 99.8% failure-point accuracy)
• Lesson:utopia fast and efficient address translation via hybrid restrictive flexible virtual to physica 796873cf  + (workloads=single-core evaluation described … workloads=single-core evaluation described by the preserved official abstract; baselines=the baseline system and the best prior contiguity-aware translation scheme; metrics=performance; results=Utopia improves performance by 24% over baseline, while the prior scheme improves it by 13%; detailed workload, page-walk, row-buffer, area, and power figures require the exact full text and are not asserted here.exact full text and are not asserted here.)
• Lesson:loom efficient capture and querying of high frequency telemetry bbcbd611  + (workloads=single-host Redis/eBPF motivatin … workloads=single-host Redis/eBPF motivating scenario and two real-world high-frequency telemetry workloads; baselines=InfluxDB, FishStore, and raw-file capture; metrics=ingest rate, data-drop rate, query latency, CPU/memory probe effect; results=Loom sustains up to 9M records/s without dropping data. InfluxDB drops 38–93% and answers queries 7–160× slower; versus FishStore, Loom improves query latency 1.5–17× while exceeding ingest performance. Its ingest and probe effect are comparable to raw-file capture.effect are comparable to raw-file capture.)
• Lesson:extending applications safely and efficiently 7a5f9f65  + (workloads=six application-extension use cases; baselines=native/in-process and isolated extension approaches; metrics=safety and efficiency; results=six use cases demonstrated; exact figures unavailable in abstract)
• Lesson:pegasus tolerating skewed workloads in distributed storage with in network coherence directories a4e8c8ec  + (workloads=skewed KV workloads; dynamic hot keys; baselines=partitioned distributed storage; metrics=throughput under latency SLO; results=>10× throughput across tested conditions.)
• Lesson:lithos an operating system for efficient machine learning on gpus 76d95512  + (workloads=stacked inference and hybrid ML workloads; baselines=NVIDIA MPS and prior GPU schedulers; metrics=p99 latency, throughput, capacity, energy; results=13x/3x lower p99 and 1.6x throughput; hybrid 4.7x/1.18x p99 and 1.35x throughput)
• Lesson:halfmoon log optimal fault tolerant stateful serverless computing e44c470a  + (workloads=stateful serverless workloads; baselines=Boki; metrics=latency; logging overhead; results=20–40% lower latency; 1.5–4× lower logging overhead.)
• Lesson:byteexpress a high performance and traffic efficient inline transfer of small payloads over nvme a7c01a49  + (workloads=sub-page NVMe payload microbenchmarks on Linux/OpenSSD; baselines=PRP and prior small-I/O method; metrics=PCIe traffic and latency; results=up to 98% less traffic; 40% and 39% lower latency)
• Lesson:a comprehensive analysis of superpage management mechanisms and policies c34f2dbd  + (workloads=superpage-management workloads; baselines=existing superpage policies; metrics=latency spikes, performance, memory bloat; results=qualitative improvement; exact values unavailable in official abstract)
• Lesson:scaleswap a scalable os swap system for all flash swap arrays 4041bd54  + (workloads=swap workloads on 128 cores and 8 NVMe SSDs; baselines=Linux, TMO, ExtMEM; metrics=throughput and average latency; results=up to 3.4x throughput and 11.5x lower latency vs Linux; +64% vs TMO; up to 5x ExtMEM)
• Lesson:the design and implementation of a capacity variant storage system 91c1817e  + (workloads=synthetic and real workloads; baselines=fixed-capacity storage; metrics=latency; throughput; lifetime; results=Latency -8–53%; throughput +49–316%; lifetime +268–327%.)
• Lesson:constructing and analyzing the lsm compaction design space 1aa4ace5  + (workloads=ten instantiated compaction stra … workloads=ten instantiated compaction strategies; baselines=representative leveling, tiering, and hybrid policies; metrics=write amplification; write throughput; point lookup; range lookup; space amplification; results=12 empirical observations; seven design takeaways; no universal winnereven design takeaways; no universal winner)
• Lesson:citadel rethinking memory allocation to safeguard against inter domain rowhammer exploits 8fd39aca  + (workloads=thousands of memory security domains; baselines=prior guard/isolation allocation; metrics=memory overhead and performance; results=7.2% average overhead, no performance loss, 4–6x less overhead than prior isolation)
• Lesson:cacheblend fast large language model serving for rag with cached knowledge fusion 056a5f5e  + (workloads=three open-source LLMs; four QA/ … workloads=three open-source LLMs; four QA/RAG/summarization datasets; baselines=full KV recomputation and naive full reuse; metrics=quality, TTFT, throughput, recomputed-token share; results=<15% typical recomputation; 2.2–3.3x TTFT and 2.8–5x throughput improvement2–3.3x TTFT and 2.8–5x throughput improvement)
• Lesson:tiered memory management beyond hotness 6ad152df  + (workloads=tiered-memory application suite; baselines=four state-of-the-art tiering systems; metrics=application performance and worst-case regression; results=up to 12.4x improvement; at most 3% underperformance)
• Lesson:pet proactive demotion for efficient tiered memory management e2185da3  + (workloads=tiered-memory workloads on Linux 6.1.44; baselines=page-based tiering; metrics=fast-memory footprint and performance; results=39.8% average/80.4% maximum footprint reduction at 1.7% average loss; mitigates 31% cliff)
• Lesson:understanding and mitigating covert and side channel vulnerabilities introduced by rowhammer def 0e272653  + (workloads=two constructed covert channels … workloads=two constructed covert channels and a website-fingerprinting attack; baselines=two state-of-the-art RowHammer defenses and three proposed countermeasures; metrics=channel capacity and mitigation trade-offs; results=the preserved arXiv abstract reports 39.0 and 48.7 Kbps channel capacity; detailed attack-success and mitigation-overhead values require the exact full-text version.alues require the exact full-text version.)
• Lesson:kvaccel a novel write accelerator for lsm tree based kv stores with host ssd collaboration 968f5966  + (workloads=write-intensive and mixed read/write key-value workloads; baselines=ADOC and conventional paths; metrics=throughput and CPU efficiency; results=up to 17% throughput gain on write-intensive workloads; comparable mixed-workload performance)
• Lesson:mitigating resource usage dependency in sorting based kv stores on hybrid storage devices via op 31cbb8ad  + (workloads=write-intensive and read-intensi … workloads=write-intensive and read-intensive key-value workloads; baselines=RocksDB, MatrixKV, PrismDB, SplitDB, ADOC; metrics=CPU utilization, throughput, tail latency; results=+25.4–32.3% CPU utilization; 2.3–4.9x write throughput; 74.3–91.4% lower tail; 1.2–2.3x read throughput91.4% lower tail; 1.2–2.3x read throughput)
• Lesson:wasp workload aware self replicating page tables for numa servers cd871a6f  + (workloads=x86 and ARM NUMA workloads; baselines=no replication; manual page-table self-replication; metrics=application performance; replication overhead; results=At least the performance improvement of manual PTSR; exact aggregate not abstract-verified.)

• Lesson:technical review memory deduplication in mobile systems 9c6aec68  + (검증 가능한 결과 없음.)

• Lesson:technical review bperf bcoz on cpu off cpu 7ecf409a  + (공개 메타데이터와 목차에는 독립적인 정량 평가 결과가 제시되지 않는다.)
• Lesson:bperf bcoz on cpu off cpu 3b259a81  + (공개 메타데이터와 목차에는 독립적인 정량 평가 결과가 제시되지 않는다.)
• Lesson:technical review e0205794  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:technical review filter lsm tree 4641bf5f  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:technical review ufs a8ea798b  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:filter lsm tree 89fd8853  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:lesson 907f0c60  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:ufs bbe26b78  + (공개된 공식 초록 범위에서는 정량 결과를 확인할 수 없다.)
• Lesson:technical review transparent smartphone memory expansion f1adb2b8  + (공식 초록 기준 displayed time 41% 감소. 메모리 용량 62% 확장과 launch time 42% 단축을 보고하며 전통적 swapping 및 app-killing보다 우수하다고 평가.)
• Lesson:transparent smartphone memory expansion 5ad5fd6a  + (공식 초록 기준 displayed time 41% 감소. 메모리 용량 62% 확장과 launch time 42% 단축을 보고하며 전통적 swapping 및 app-killing보다 우수하다고 평가.)
• Lesson:technical review nvme driven lazy cache coherence for immutable data with nvme over fabrics 4d6bc037  + (공식 초록은 TCP/IP remote read와 compute-storage node 간 network traffic을 줄여 MinIO 성능을 개선했다고 보고한다. 공개 초록에는 개선율·latency·throughput 수치가 없다.)
• Lesson:nvme driven lazy cache coherence for immutable data with nvme over fabrics c19688e2  + (공식 초록은 TCP/IP remote read와 compute-storage node 간 network traffic을 줄여 MinIO 성능을 개선했다고 보고한다. 공개 초록에는 개선율·latency·throughput 수치가 없다.)
• Lesson:technical review efficient memory deduplication for mobile smart devices 6603ecf1  + (공식 초록은 기존 접근과 같은 메모리 절감을 유지하면서 계산비용을 유의하게 줄였다고 보고한다. 정량값은 공개 초록에 없다.)
• Lesson:efficient memory deduplication for mobile smart devices fa9b72e9  + (공식 초록은 기존 접근과 같은 메모리 절감을 유지하면서 계산비용을 유의하게 줄였다고 보고한다. 정량값은 공개 초록에 없다.)
• Lesson:technical review improving application launch performance in smartphones using recurrent neural 47c597b7  + (공식 초록은 대부분의 앱을 background에 준비시키고 launch latency를 유의하게 줄였다고 보고하지만 수치를 제공하지 않는다.)
• Lesson:improving application launch performance in smartphones using recurrent neural network 8cfcaeb5  + (공식 초록은 대부분의 앱을 background에 준비시키고 launch latency를 유의하게 줄였다고 보고하지만 수치를 제공하지 않는다.)
• Lesson:technical review bdff2f7c  + (공식 초록은 제안 기법의 설계와 구현을 확인하지만 정량 startup 개선치는 공개하지 않는다.)
• Lesson:lesson 13ed6bae  + (공식 초록은 제안 기법의 설계와 구현을 확인하지만 정량 startup 개선치는 공개하지 않는다.)
• Lesson:technical review lsm tree da534157  + (공식 초록은 제안 기법이 LSM-tree 읽기 경로의 직렬 대기를 줄인다고 설명하지만 정량 수치는 공개하지 않는다.)
• Lesson:lsm tree 060db792  + (공식 초록은 제안 기법이 LSM-tree 읽기 경로의 직렬 대기를 줄인다고 설명하지만 정량 수치는 공개하지 않는다.)
• Lesson:technical review a scalable and overflow tolerant mechanism for minimum virtual time tracking 7a68da61  + (공식 초록은 최신 공정 I/O 스케줄러 통합 시 overflow 이상 없이 공정성과 안정 실행을 유지했다고 보고한다. 공개 초록에는 성능/오버헤드 수치가 없다.)
• Lesson:a scalable and overflow tolerant mechanism for minimum virtual time tracking b9b92703  + (공식 초록은 최신 공정 I/O 스케줄러 통합 시 overflow 이상 없이 공정성과 안정 실행을 유지했다고 보고한다. 공개 초록에는 성능/오버헤드 수치가 없다.)
• Lesson:technical review memsos os guided selective memory mirroring 4aec983d  + (기존 최신 partial mirroring 대비 시스템 FIT를 최대 19,000배 낮춤. 성능 오버헤드 3% 미만이며, 여러 경우 full mirroring에 근접한 신뢰성을 보고.)
• Lesson:memsos os guided selective memory mirroring ac9aab48  + (기존 최신 partial mirroring 대비 시스템 FIT를 최대 19,000배 낮춤. 성능 오버헤드 3% 미만이며, 여러 경우 full mirroring에 근접한 신뢰성을 보고.)
• Lesson:dwkv metric mismatch  + (단일 corrected seed에서 DW-KV-order-only는 valu … 단일 corrected seed에서 DW-KV-order-only는 value-weighted tardiness 2398 ms로 최저였지만 premium miss 36.4%였고, tier-priority와 length-aware는 premium miss 12.1%였다. DW-KV premium E2E p90은 약 33.1 s, tier-priority는 약 6.0 s였다. 같은 run에서 KV denominator 기여는 value-urgency 2492→DW-KV 2398 ms, 3.8%였으며 single-seed/unverified였다.2398 ms, 3.8%였으며 single-seed/unverified였다.)