본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

고급검색

인명/단체명

저자정보 상세정보
인명/단체명을 입력하세요.

키워드

  • 대표어

  • 외국어

-

목차보기

[표지] 1

제출문 2

보고서 초록 3

요약문 4

Summary 7

목차 11

CONTENTS 12

제1장 사업 개요 19

제1절 목적 및 배경 19

1. 목적 19

2. 배경 19

제2절 연구사업의 중요성 21

1. 4차 산업 혁명과 초고성능 컴퓨터 21

2. 초고성능컴퓨팅과 데이터과학의 발전 22

3. 초고성능컴퓨팅과 제조업 디지털 혁신 23

제3절 연구사업의 필요성 24

제4절 사업목표 25

1. 최종 사업 목표 25

2. 사업 내용 25

3. 대내외 추진 체계도 26

4. 연구사업 추진전략 및 방법 27

제5절 사업 동향 28

1. 국내 동향 28

2. 국외 동향 31

제2장 사업 성과 35

제1절 성과지표 및 달성도 35

1. 성과 지표별 달성도 종합 35

2. 대표성과 36

제2절 연구사업 세부 성과 39

1. 슈퍼컴퓨터 인프라 구축 및 운영 39

2. 슈퍼컴퓨팅 일반 사용자 지원 75

3. 지적재산권 98

제3절 연구사업 결과의 기대성과 및 활용방안 100

부록 1. 2020년 슈퍼컴퓨터 운영 현황 통계 102

[뒷표지] 117

List of Tables 13

〈Table I-1〉 Main functions of national supercomputing center 19

〈Table I-2〉 Council on Competitiveness(USA, '15) 23

〈Table I-3〉 The study target for every year and the final 25

〈Table I-4〉 The strategies and methods 27

〈Table I-5〉 Exampes of big supercomputing study projects 28

〈Table I-6〉 The amount of supercomputers and performance in domestic 30

〈Table I-7〉 The world supercomputer ranking(Nov. 2020) 30

〈Table I-8〉 Establishment of inter-relationship for using supercomputers 30

〈Table I-9〉 The top nations competition for supercomputer power 31

〈Table II-1〉 Summary of performance indicators 35

〈Table II-2〉 Stabilization Technology and Service Improvements of KISTI-5 supercomputer(Nurion) 36

〈Table II-3〉 Comparison of MTBI between KISTI-5 supercomputer(Jan.~Oct. 2020) and major TOP 20 supercomputers 37

〈Table II-4〉 Supercomputer Infrastructure Status 39

〈Table II-5〉 The main specifications for 5th supercomputer[이미지참조] 40

〈Table II-6〉 PSF Server configuration 44

〈Table II-7〉 IME Single node configuration 45

〈Table II-8〉 Burst buffer full configuration 46

〈Table II-9〉 S/W upgrade list of KISTI-5 supercomputer(nurion) 47

〈Table II-10〉 UNIX server diagnostics result of KISTI-5 supercomputer(Nurion) 69

〈Table II-11〉 Information system diagnostics result of KISTI-5 supercomputer(Nurion) 69

〈Table II-12〉 Network equipment diagnostics result of KISTI-5 supercomputer(Nurion) 70

〈Table II-13〉 Security vulnerability analysis result and protection measure status 70

〈Table II-14〉 Number of technical support cases for supercomputer user 77

〈Table II-15〉 Software installation status of Nurion system 79

〈Table II-16〉 Software installation status of Neuron system 79

〈Table II-17〉 Number of information postings on how to install SW and other notice on the... 80

List of Figures 15

〈Figure I-1〉 The roles of HPC on the intelligency and information society 21

〈Figure I-2〉 Examples of HPC 22

〈Figure I-3〉 Apply HPC to manufacturing 23

〈Figure I-4〉 Promotion system for in- or external relationships 26

〈Figure II-1〉 Usage rate and service availability(Jan.~Oct. 2020) of KISTI-5 supercomputer(Nurion) 37

〈Figure II-2〉 Green IDC Platinum Certificate, Jan. 2020 38

〈Figure II-3〉 System diagram of 5th supercomputer[이미지참조] 40

〈Figure II-4〉 Interconnection network system 41

〈Figure II-5〉 Ranking of KISTI-5 supercomputer(Nurion) in the TOP500 list(Nov. 2020) 41

〈Figure II-6〉 Constellation of KISTI-5 supercomputer(Nurion) in the main system room 42

〈Figure II-7〉 Storage systems of KISTI-5 supercomputer(Nurion) 42

〈Figure II-8〉 Parallel file system and burst buffer configuration 43

〈Figure II-7〉 Burst buffer function 45

〈Figure II-8〉 Burst buffer server configuration 45

〈Figure II-9〉 Comparison of applications execution performance before and after kernel update 48

〈Figure II-10〉 MCDRAM bandwidth comparison of depending on how the application runs 49

〈Figure II-11〉 Status check and automatic recovery procedures of computing nodes 49

〈Figure II-12〉 Backfill scheduling policy example 51

〈Figure II-13〉 Comparison of average latency by the number of nodes before and after... 52

〈Figure II-14〉 Lustre upgrade procedure for performance improvement of parallel filesystem 53

〈Figure II-15〉 Lnet router installation for intercommunication of Lustre filesystem between... 53

〈Figure II-16〉 Realtime I/O statistics for ES-Mon monitoring 54

〈Figure II-17〉 Usage and I/O throughput statistics of burst buffer 55

〈Figure II-18〉 IO500 ranking of burst buffer 56

〈Figure II-19〉 All-flash system deployment for POC 56

〈Figure II-20〉 BMT result of all-flash system for POC 57

〈Figure II-21〉 Configuration diagram of Neuron system 58

〈Figure II-22〉 Application usage statistics of Neuron system 59

〈Figure II-23〉 Monthly number of user support status of Neuron system 59

〈Figure II-24〉 SW usage statistics of Neuron system(Based on CPU time) 60

〈Figure II-25〉 Usage statistics by type of AI utilization environment 60

〈Figure II-26〉 Performance comparison of mvapich2 and mvapich2-x 61

〈Figure II-27〉 Application performance analysis based on burst buffer 62

〈Figure II-28〉 Data path depending on how the GPUDirect is configured 62

〈Figure II-29〉 GPU configuration status by Neuron computing node type 63

〈Figure II-30〉 P2P, B/W, NUMA effect, and application performance measurement by GPU... 64

〈Figure II-31〉 Supercomputer PUE(2020) 65

〈Figure II-32〉 Cooling flow diagram of LCP 66

〈Figure II-33〉 Modeling of supercomputer room based on EnergyPlus and CFD 67

〈Figure II-34〉 Security vulnerability analysis result of KISTI-5 supercomputer(Nurion) 68

〈Figure II-35〉 Physical vulnerability analysis result of KISTI-5 supercomputer(Nurion) 68

〈Figure II-36〉 Result report for security vulnerability analysis and protection measure 71

〈Figure II-37〉 Protection measures for major Information and Communication Infrastructure 71

〈Figure II-38〉 Implementation plan for protection measure and self-inspection evidence 72

〈Figure II-39〉 Business Continuity Plan to Prepare for the outbreak of corona-19... 72

〈Figure II-40〉 Recovery simulation training plan for supercomputer user data 73

〈Figure II-41〉 Security training materials and attendees for employees of supercomputer service company 73

〈Figure II-42〉 CCTV installation status for monitoring visitor behavior 74

〈Figure II-43〉 Satisfaction survey results of user consultation 75

〈Figure II-44〉 Main screen of SOS(Supercomputer Operation System) 75

〈Figure II-45〉 Supercomputer customer satisfaction survey results(2019) 76

〈Figure II-46〉 Supercomputer customer satisfaction survey results:VOC part(2019) 76

〈Figure II-47〉 Number of technical support cases per month 77

〈Figure II-48〉 Technical support status by type 77

〈Figure II-49〉 Supercomputer customer satisfaction survey results:VOC part(2019) 78

〈Figure II-50〉 Number of visits to the supercomputing blog per month 80

〈Figure II-51〉 AI web service configuration diagram of Neuron system based on JupyterHub 81

〈Figure II-52〉 Jupyter notebook execution procedure based on conda virtual environment 81

〈Figure II-53〉 Jupyter service diagram of Neuron system 82

〈Figure II-54〉 Resource status screen of Jupyter service 82

〈Figure II-55〉 IPyParallel screen of Neuron jupyter service 83

〈Figure II-56〉 Data pre-processing procedure diagram of geoDIP 83

〈Figure II-57〉 Job submission screen for data pre-processing of geoDIP 84

〈Figure II-58〉 VDI service procedure diagram 85

〈Figure II-59〉 Training practice examples of Jupyter notebook for deep learning 86

〈Figure II-60〉 Training practice screen of Jupyter notebook for deep learning 86

〈Figure II-61〉 Integration of multiple homepages for Supercomputer service 87

〈Figure II-62〉 Utilization status screen of Nurion system 88

〈Figure II-63〉 Various utilization statistics screen for supercomputer 88

〈Figure II-64〉 Compiler & MPI version and Optimization option combination list 89

〈Figure II-65〉 GRIMS model execution result by configuration type 90

〈Figure II-66〉 GRIMS model execution result by number of threads on hybrid... 90

〈Figure II-67〉 Deep learning application performance analysis results on Neuron and the... 91

〈Figure II-68〉 Partnership establishment between KISTI and CNU for Bio AI Convergence Research 92

〈Figure II-69〉 SARS-CoV-2 T cell epitope sequence motif and conservation analysis 93

〈Figure II-70〉 Predictive performance analysis for NetMHCpan, MixMHCpred, and MHCflurry 93

〈Figure II-71〉 The whole learning process based on transfer learning 94

〈Figure II-72〉 2D data center model 94

〈Figure II-73〉 Distribution of speed field according to various input speeds 95

〈Figure II-74〉 Distribution of temperature field according to various input speeds 95

〈Figure II-75〉 Temperature change over time at server output stage 96

〈Figure II-76〉 Supercomputer education support and promotion achievements 97

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차

권호

기사명 원문보기 기사목차
닫기