속성:Context
하드웨어, 부하, 버전, 규모처럼 결과에 영향을 줄 수 있는 조건을 적습니다.
c
Venue: EuroSys. Year: 2025.
Counter/page-fault/access-bit/PEBS approaches trade frequency resolution against overhead and can misclassify hot pages or bloat hot regions.
Verification: author-hosted 19-page full paper + official EuroSys 2025 listing; confidence=high. +
citadel rethinking memory allocation to safeguard against inter domain rowhammer exploits 8fd39aca +
Venue: MICRO. Year: 2025.
Guard-row approaches often waste substantial DRAM as the number of isolated domains grows.
Verification: official DOI/publisher abstract; confidence=high. +
cold archives must materialize content addressed model files and cross a storage failure domain 03687d8e +
B200-2 Qwen/Mistral generation attempts a01-a04, with source data on Lustre and model snapshots whose files point into a shared Hugging Face blob store. B200 access is temporary. +
Venue: FAST. Year: 2024.
작은 I/O는 버퍼링이 유리하지만 큰·경합 I/O는 direct가 유리해 정적 선택이 어렵다.
Verification: abstract_only; confidence=high. +
B200-2 a03 sequential Qwen and Mistral target-generation migration run under commit dd81c5ff98ffce72cd907a416fd42efbb41f920b. +
Publication scope: international. Lab publication metadata:
1. Jeaho Hwang, Jinkyu Jeong, Hwanju Kim, Jeonghwan Choi, and Joonwon Lee, "Compressed Memory Swap for QoS of Virtualized Embedded Systems," IEEE Transactions on Consumer Electronics, vol. 58, no. 3, pp. 834-840, Aug. 2012
Verification level: official_abstract. Sources checked:
https://doi.org/10.1109/TCE.2012.6311325
https://yonsei.elsevierpure.com/en/publications/compressed-memory-swap-for-qos-of-virtualized-embedded-systems/ +
Venue: VLDB. Year: 2021.
Compaction choices trade write, read, and space amplification, but prior systems mix several design decisions.
Verification: full_text; confidence=high. +
Publication scope: international. Lab publication metadata:
1. Sang-Hoon Kim, Sejun Kwon, Jin-Soo Kim, and Jinkyu Jeong, "Controlling Physical Memory Fragmentation in Mobile Systems," in Proceedings of International Symposium on Memory Management (ISMM 2015), Portland, OR, USA, June 13-14, 2015. - Highest ranked paper in double-blind review process
Verification level: official_abstract. Sources checked:
https://doi.org/10.1145/2754169.2754179
https://www.sigplan.org/OpenTOC/ismm15.html
https://yonsei.elsevierpure.com/en/publications/controlling-physical-memory-fragmentation-in-mobile-systems +
Venue: VLDB. Year: 2022.
전통적 2PC의 코디네이터 결정 로그가 원격 저장소 왕복과 단일 병목을 만든다.
Verification: full_text; confidence=high. +
Venue: VLDB. Year: 2021.
Modern in-memory engines are often memory-latency bound, but conventional batching complicates transaction code.
Verification: full_text; confidence=high. +
cost efficient large language model serving for multi turn conversations with cachedattention aa200b5c +
Venue: USENIX ATC. Year: 2024.
Recomputing the conversation prefix wastes prefill work; retaining every KV cache on GPU limits concurrency.
Verification: official_abstract; confidence=high. +
Publication scope: international. Lab publication metadata:
1. Keun Soo Lim, Yunjay Hong, Jongheon Jeong, Sam Son, Donguk Kim, Yeonhong Park, Jae W. Lee, Jinkyu Jeong, "CPC: Coordinated Page Cache for Serverless Computing," in Proceedings of The 34th International Conference on Parallel Architectures and Compilation Techniques (PACT '25), Nov 3 - 6, Irvine, California, US
Verification level: official_abstract. Sources checked:
https://yonsei.elsevierpure.com/en/publications/cpc-coordinated-page-cache-for-serverless-computing/
https://doi.org/10.1109/PACT65351.2025.00019
https://pact2025.github.io/program +
Venue: ASPLOS. Year: 2024.
OS-only prefetchers see block streams but not application intent; demand I/O can also block useful prefetches.
Verification: official_abstract; confidence=high. +
d
Publication scope: international. Lab publication metadata:
1. Jiwon Woo, Minwoo Ahn, Gyusun Lee, and Jinkyu Jeong "D2FQ: Device-Direct Fair Queueing for NVMe SSDs," in Proceedings of 19th USENIX Conference on File and Storage Technologies (FAST'21), Virtual event, Feb 23–25, 2021. slides, video
Verification level: full_text. Sources checked:
https://www.usenix.org/conference/fast21/presentation/woo
https://www.usenix.org/system/files/fast21-woo.pdf +
Publication scope: international. Lab publication metadata:
1. Jinkyu Jeong, Hwanju Kim, Jeaho Hwang, Joonwon Lee, and Seungryoul Maeng, "DaaC: Device-reserved Memory as an Eviction-based File Cache," the 21st International Conference on Compilers Architecture and Synthesis for Embedded Systems (CASES 2012), Tampere, Finland, October, 2012
Verification level: official_abstract. Sources checked:
https://doi.org/10.1145/2380403.2380439
https://yonsei.elsevierpure.com/en/publications/daac-device-reserved-memory-as-an-eviction-based-file-cache/
https://koasas.kaist.ac.kr/handle/10203/175991 +
Venue: EuroSys. Year: 2025.
Fine-tuned models share a base lineage but conventional runtimes duplicate weights and serialize costly model swaps.
Verification: arXiv full-text excerpts, DOI metadata, and official acceptance page; confidence=high. +
Publication scope: international. Lab publication metadata:
1. Hwanju Kim, Sangwook Kim, Jinkyu Jeong, Joonwon Lee, and Seungryoul Maeng, "Demand-Based Coordinated Scheduling for SMP VMs," in Proceedings of the 18th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS XVIII), Houston, Texas, USA, March, 2013
2. Hwanju Kim, Sangwook Kim, Jinkyu Jeong, Joonwon Lee, and Seungryoul Maeng, "Demand-based Coordinated Scheduling for SMP VMs," ACM SIGPLAN Notices, vol. 48, no. 4, pp. 369-380, Apr. 2013
Verification level: official_abstract. Sources checked:
https://doi.org/10.1145/2451116.2451156
https://doi.org/10.1145/2499368.2451156
https://yonsei.elsevierpure.com/en/publications/demand-based-coordinated-scheduling-for-smp-vms/ +
Publication scope: international. Lab publication metadata:
1. Seonyeong Park, Jinkyu Jeong, Heeseung Jo, Joonwon Lee, and Euiseong Seo, "Development of Behavior-Profilers for Multimedia Consumer Electronics," IEEE Transactions on Consumer Electronics, Vol. 55, No. 4, pp. 1929-1935, November 2009
Verification level: official_abstract. Sources checked:
https://doi.org/10.1109/TCE.2009.5373752
https://yonsei.elsevierpure.com/en/publications/development-of-behavior-profilers-for-multimedia-consumer-electro/ +
Publication scope: international. Lab publication metadata:
1. Hyeonjun An, Sihyun Kim, Chaerim Lim, Hyunjoon Kim, Rathijit Sen, Sangmin Jung, Hyeonsoo Lee, Dongwook Kim, Takki Yu, Jinkyu Jeong, Youngsok Kim, and Kwanghyun Park, "DFLOP: A Data-driven Framework for Multimodal LLM Training Pipeline Optimization," in Proceedings of the International Conference on Management of Data (SIGMOD '26), 31 May 2026- 5 June 2026, Bengaluru India
Verification level: full_text. Sources checked:
https://arxiv.org/abs/2603.25120
https://arxiv.org/pdf/2603.25120
https://www.microsoft.com/en-us/research/publication/dflop-a-data-driven-framework-for-multimodal-llm-training-pipeline-optimization/ +
Venue: ASPLOS. Year: 2023.
Centralized RAID controllers and blocking multistage writes underuse peer bandwidth and magnify network overhead.
Verification: official_abstract; confidence=high. +