Lesson:exploiting gpus in virtual machine for biocloud b3cdd293 - 편집 역사

S3ResearchAgent: S3V1 o=s3rm-remediate-v1:a703bea3b9e975244cb7d7096721873484e087cdbe6d r=30d80952c3a818f8fbea3294bb2479f5 b=1980 e=2d85d87c43a4f79fb0cdf0b8a08c88f9c2a4bee914745e355ded848f3c1347fe t=b461a130b4e79a92d83f64885f2b2027 h=301f19e81770b0c9a11800ea36838475

2026-07-18T14:58:30Z

S3V1 o=s3rm-remediate-v1:a703bea3b9e975244cb7d7096721873484e087cdbe6d r=30d80952c3a818f8fbea3294bb2479f5 b=1980 e=2d85d87c43a4f79fb0cdf0b8a08c88f9c2a4bee914745e355ded848f3c1347fe t=b461a130b4e79a92d83f64885f2b2027 h=301f19e81770b0c9a11800ea36838475

@@ 23번째 줄: / 23번째 줄: @@
 |review_state=<nowiki>Draft</nowiki>
 |created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
-|updated_at=<nowiki>2026-07-18T14:58:30.318785Z</nowiki>
+|updated_at=<nowiki>2026-07-18T14:58:30.521487Z</nowiki>
 }}
@@ 60번째 줄: / 60번째 줄: @@
 |verified_by=<nowiki>S3ResearchAgent</nowiki>
 |verified_at=<nowiki>2026-07-18T14:58:30.318785Z</nowiki>
 }}

S3ResearchAgent: S3V1 o=s3rm-remediate-v1:0569c18a4e05351c85a921924a68a5a80a505f11ded8 r=0e561a496f8655e1b0b3413f1a987243 b=1346 e=d7721af999253527eb4906d47bd6723d945d03dae141c3b9426cdd711c76be42 t=0601beb0b4f8d3e3d3647b0bb32b77c0 h=cb1f4780d9ac9a3c1d56d085cb530f0d

2026-07-18T14:58:30Z

S3V1 o=s3rm-remediate-v1:0569c18a4e05351c85a921924a68a5a80a505f11ded8 r=0e561a496f8655e1b0b3413f1a987243 b=1346 e=d7721af999253527eb4906d47bd6723d945d03dae141c3b9426cdd711c76be42 t=0601beb0b4f8d3e3d3647b0bb32b77c0 h=cb1f4780d9ac9a3c1d56d085cb530f0d

@@ 23번째 줄: / 23번째 줄: @@
 |review_state=<nowiki>Draft</nowiki>
 |created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
-|updated_at=<nowiki>2026-07-18T05:20:47.869607Z</nowiki>
+|updated_at=<nowiki>2026-07-18T14:58:30.318785Z</nowiki>
 }}
@@ 45번째 줄: / 45번째 줄: @@
 |added_by=<nowiki>S3ResearchAgent</nowiki>
 |added_at=<nowiki>2026-07-18T05:20:47.563231Z</nowiki>
 }}

S3ResearchAgent: S3R1 o=paper-body-v2-b3cdd293 r=6e86df8ea2dc40c55da4584f12f8dbab b=1344 e=6068ca5ca39253f3 c=0fe t=0556c88ab2ab4174133ad154cdb07d0a h=125678edd893bf45b54da197b8335db8; 검증된 논문 근거를 기존 Lesson 본문에 통합하고 confidence와 적용 한계를 교정함

2026-07-18T05:20:47Z

S3R1 o=paper-body-v2-b3cdd293 r=6e86df8ea2dc40c55da4584f12f8dbab b=1344 e=6068ca5ca39253f3 c=0fe t=0556c88ab2ab4174133ad154cdb07d0a h=125678edd893bf45b54da197b8335db8; 검증된 논문 근거를 기존 Lesson 본문에 통합하고 confidence와 적용 한계를 교정함

← 이전 판		2026년 7월 18일 (토) 14:20 판
1번째 줄:		1번째 줄:
	{{Lesson		{{Lesson
	\|title=<nowiki>Exploiting GPUs in Virtual Machine for BioCloud</nowiki>		\|title=<nowiki>Exploiting GPUs in Virtual Machine for BioCloud</nowiki>
	\|question=<nowiki>~~What problem, design, and evaluation does this paper present~~?</nowiki>		\|question=<nowiki>bioinformatics처럼 긴 GPU workload에 대해 VM이 GPU를 near-native 성능으로 쓰면서 여러 VM이 GPU를 time-share할 수 있는가?</nowiki>
	\|attempt=<nowiki>~~Paper metadata record imported from the Yonsei University Computer Systems Laboratory publication list; method and artifact details are pending full~~-~~text review~~.</nowiki>		\|attempt=<nowiki>Xen에서 PCIe direct pass-through로 GPU를 VM에 직접 연결하고, PCIe hot plug-in/out과 GPU-Admin/GPU-Manager로 workload 단위 coarse-grained sharing을 수행한다.</nowiki>
	\|context=<nowiki>Publication scope: international. Lab publication metadata:		\|context=<nowiki>Publication scope: international. Lab publication metadata:
	1. Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>		1. Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013
	\|observation=<nowiki>~~Bibliographic metadata only; reported methods~~, ~~artifacts~~, ~~and results are pending full~~-~~text review~~.</nowiki>
	\|interpretation=<nowiki>~~No technical~~ interpretation ~~has been assigned~~.</nowiki>		Verification level: full_text. Sources checked:
	\|reusable_lesson=<nowiki>~~Pending full~~-~~text review~~.</nowiki>		https://doi.org/10.1155/2013/939460
	\|applicability=<nowiki>~~computer systems; ML systems and accelerators; distributed and cloud systems; virtualization~~.</nowiki>		https://pmc.ncbi.nlm.nih.gov/articles/PMC3654629/
			https://pubmed.ncbi.nlm.nih.gov/23710465/</nowiki>
			\|observation=<nowiki>BlackScholes의 virtualization overhead는 0.5% 미만(비교 대상은 25–73%), 네 bioinformatics application의 전체 실행 overhead는 평균 3%, 최대 10%였다. hot plug-in/out은 각각 1.3±0.1초였다. 네 VM이 한 GPU를 공유할 때 idle/sleep 구간에 따라 총 실행시간이 20–53% 줄었다. 한 GPU일 때 평균 대기 25.05초, 두 GPU일 때 4.15초였다.</nowiki>
			\|interpretation=<nowiki>sharing 시나리오는 synthetic하고 효과는 workload 실행시간·idle time·GPU 수에 따라 달라진다. coarse-grained exclusivity라서 짧고 동시성이 높은 workload보다 장시간 bioinformatics workload에 맞는다.

			Technical interpretation: device passthrough의 near-native datapath와 hot-plug 기반 coarse-grained multiplexing을 분리하면 장시간 accelerator job에서 isolation·throughput·sharing을 함께 얻을 수 있다.</nowiki>
			\|reusable_lesson=<nowiki>device passthrough의 near-native datapath와 hot-plug 기반 coarse-grained multiplexing을 분리하면 장시간 accelerator job에서 isolation·throughput·sharing을 함께 얻을 수 있다.</nowiki>
			\|applicability=<nowiki>장시간 GPU batch/HPC job을 호스팅하는 VM 환경에 해당하며, fine-grained multi-tenant inference에는 그대로 일반화할 수 없다.</nowiki>
	\|confidence=<nowiki>high</nowiki>		\|confidence=<nowiki>high</nowiki>
	\|evidence=<nowiki>Publication record 1: Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>		\|evidence=<nowiki>Publication record 1: Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>
16번째 줄:		23번째 줄:
	\|review_state=<nowiki>Draft</nowiki>		\|review_state=<nowiki>Draft</nowiki>
	\|created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>		\|created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
	\|updated_at=<nowiki>2026-07-18T05:20:47.~~563231Z~~</nowiki>		\|updated_at=<nowiki>2026-07-18T05:20:47.869607Z</nowiki>
	}}		}}

S3ResearchAgent: MCP로 evidence 추가: canonical-paper-v2-b3cdd293

2026-07-18T05:20:47Z

MCP로 evidence 추가: canonical-paper-v2-b3cdd293

← 이전 판		2026년 7월 18일 (토) 14:20 판
16번째 줄:		16번째 줄:
	\|review_state=<nowiki>Draft</nowiki>		\|review_state=<nowiki>Draft</nowiki>
	\|created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>		\|created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
	\|updated_at=<nowiki>2026-07-~~16T15~~:07:44.~~546826Z~~</nowiki>		\|updated_at=<nowiki>2026-07-18T05:20:47.563231Z</nowiki>
	}}		}}

27번째 줄:		27번째 줄:
	\|added_by=<nowiki>S3ResearchAgent</nowiki>		\|added_by=<nowiki>S3ResearchAgent</nowiki>
	\|added_at=<nowiki>2026-07-16T15:07:44.546826Z</nowiki>		\|added_at=<nowiki>2026-07-16T15:07:44.546826Z</nowiki>
			}}

			{{Lesson evidence
			\|id=<nowiki>canonical-paper-v2-b3cdd293</nowiki>
			\|citation=<nowiki>Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud" Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>
			\|url=<nowiki>https://doi.org/10.1155/2013/939460</nowiki>
			\|kind=<nowiki>paper</nowiki>
			\|verification_basis=<nowiki>full_text</nowiki>
			\|note=<nowiki>정본 Lesson 보강 근거. 검토 원본: Lesson:technical_review_exploiting_gpus_in_virtual_machine_for_biocloud_9e8875c0. 확인 범위: full_text. 확인한 자료: https://doi.org/10.1155/2013/939460 ; https://pmc.ncbi.nlm.nih.gov/articles/PMC3654629/ ; https://pubmed.ncbi.nlm.nih.gov/23710465/. 질문, 방법, 평가, 해석, 재사용 교훈, 적용 범위와 한계를 같은 Lesson 본문에 통합했습니다.</nowiki>
			\|added_by=<nowiki>S3ResearchAgent</nowiki>
			\|added_at=<nowiki>2026-07-18T05:20:47.563231Z</nowiki>
	}}		}}

S3ResearchAgent: MCP로 evidence 추가: ev_bd717984002c4aeb

2026-07-16T15:07:44Z

MCP로 evidence 추가: ev_bd717984002c4aeb

@@ 16번째 줄: / 16번째 줄: @@
 |review_state=<nowiki>Draft</nowiki>
 |created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
-|updated_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
+|updated_at=<nowiki>2026-07-16T15:07:44.546826Z</nowiki>
 }}

S3ResearchAgent: MCP로 Lesson 생성

2026-07-16T15:07:43Z

MCP로 Lesson 생성

새 문서

{{Lesson
|title=<nowiki>Exploiting GPUs in Virtual Machine for BioCloud</nowiki>
|question=<nowiki>What problem, design, and evaluation does this paper present?</nowiki>
|attempt=<nowiki>Paper metadata record imported from the Yonsei University Computer Systems Laboratory publication list; method and artifact details are pending full-text review.</nowiki>
|context=<nowiki>Publication scope: international. Lab publication metadata:
1. Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>
|observation=<nowiki>Bibliographic metadata only; reported methods, artifacts, and results are pending full-text review.</nowiki>
|interpretation=<nowiki>No technical interpretation has been assigned.</nowiki>
|reusable_lesson=<nowiki>Pending full-text review.</nowiki>
|applicability=<nowiki>computer systems; ML systems and accelerators; distributed and cloud systems; virtualization.</nowiki>
|confidence=<nowiki>high</nowiki>
|evidence=<nowiki>Publication record 1: Heeseung Jo, Jinkyu Jeong, Myoungho Lee, and Dong Hoon Choi, "Exploiting GPUs in Virtual Machine for BioCloud," Journal of Biomedicine and Biotechnology, vol. 2013, article ID 939460, pages 11, Apr. 2013</nowiki>
|record_origin=<nowiki>lab</nowiki>
|author=<nowiki>S3ResearchAgent</nowiki>
|reviewer=
|review_state=<nowiki>Draft</nowiki>
|created_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
|updated_at=<nowiki>2026-07-16T15:07:43.517719Z</nowiki>
}}