CLoud System with applied Artificial iNtelligence Lab
Welcome to Cloud Systems with Applied Artificial Intelligence Lab at KNU
경북대학교 클라우드 시스템 응용AI 연구실
Director: Professor Byungchul Tak (탁병철 교수)
● Member Lab of ERC, The Engineering Research Center, Year 2021~2028 (선도연구센터지원사업 공학분야 참여 연구실, 2021년~2028년)
● KNU CSE BK21 FOUR Project Member Lab (경북대 컴퓨터학부 4단계 BK21 사업 참여 연구실)
The mission of our lab is to develop advanced techniques for managing a large number of distributed system nodes in a cloud environment. Our research area includes:
Log Analytics using deep learning techniques, mostly for large-scale system problem diagnosis.
Cloud security, container security.
Fog/Edge computing.
System resource monitoring for accounting and performance management.
We strive to develop techniques that help run distributed software components in distributed systems made of a large number of computing nodes. These techniques start from understanding the system states via theoretical and empirical analysis. Then, we actively borrow state-of-the-art techniques from other fields and apply them to our problems. Therefore, we need to have strong background knowledge not only in the operating system, computer architectures, and distributed systems, but also in data mining, and AI(Artificial Intelligence), as well as DL(Deep Learning) domains. Data generated from large-scale cloud systems are overwhelming for humans to manually process. We must have an automated data processing pipeline to be able to comprehend these machine-generated data.
Current research in our lab includes diverse topics related to cloud computing systems. We especially focus on scalable and stable distributed systems and application solutions. Specific research topics are:
In the log analytics:
System problem diagnosis using BERT model (from NLP).
Log observation via fault injection
In cloud security and container security:
Quantification of security strength of container runtimes
System call sequence-based container defense
In Fog/Edge Computing:
Query execution optimization in the fog nodes
In System Resource and Performance Monitoring/Management
NoSQL database resource monitoring and shared resource accounting
In addition to these, we work on the following closely related fields: Big data processing, Query Processing, Data Mining, NLP, Statistics, Artificial Intelligence, Deep Learning, Data Visualization, and Information Retrieval.
(추가적인 프로젝트 내용은 Research 페이지 참조.)
Announcements
클라우드 컴퓨터 시스템 분야에 관심이 많고, 최신의 인공지능 기법을 습득하여 깊이 있는 연구를 하고자 하는 석사/박사 과정 학생을 모집합니다.
bctak at knu.ac.kr로 이메일 컨택, 예약 후 연구실 방문.
석사 후 취업 또는 박사 유학을 고려하는 학생은 그에 맞춘 집중 지도.
연구실로부터의 다양한 지원을 통해 금전적 부담없이 학업과 연구에 집중 가능.
클라우드 및 대규모 시스템 분야의 연구를 수행할 석/박사 과정 학생을 모집합니다. 인공지능 분야의 기술들이 빠르게 발전함에 따라 인공지능/딥러닝 기법을 사용하여 클라우드 시스템의 문제를 해결하는 연구 방향이 크게 주목받고 활기를 띄고 있습니다. CLEAN Lab에서는 인공지능/딥러인 기법 자체의 개발이 아닌, 이를 사용하여 실질적인 클라우드 시스템의 문제들을 해결하는 것이 초점을 맞추고 있습니다. 현존하는 최신 인공지능/딥러닝 기법을 깊이있게 공부하여 클라우드에 적용함으로써 우수한 연구 결과를 내고자 하는 학생들의 지원을 바랍니다.
현재 업계에서는 대규모 클라우드 시스템을 기반으로 하는 서비스들이 빠른 속도로 발전하고 있어 이 분야의 취업기회가 전체 IT분야에서 약 40%에 달하는 것으로 알려져있습니다. 인공지능 자체의 기법에 대한 연구보다는 인공지능/딥러닝 기법을 실제 문제에 적용할 수 있는 능력을 가진 인력에 대한 많은 수요가 있습니다. 졸업 시 높은 경쟁력 확보를 위해 지금 연구에서도 클라우드의 분석, 보안 등에 딥러닝 기법, NLP(자연어처리)기법, 정보이론기법, 빅데이터처리 기법 등을 적극적으로 적용하여 연구를 수행하고 있습니다.
News
2023/11/8: 석사 졸업생 송소민 현대자동차 최종합격! Somin Song received an offer from Hyundai Motors, Korea. Congratulations!!!
2023/8/5: Our paper, "MicroVM on Edge: Is It Ready for Prime Time?" is accepted to IEEE MASCOTS'23 conference. Congratulations!!
2023/6/1: Our paper, "ECM: An Energy-efficient HVAC Control Framework for Stable Construction Environment" is accepted to IEEE SECON'23 conference. Congratulations!!
2023/5/10: Our paper, "On the Value of Sequence-based System Call Filtering for Container Security" is accepted to IEEE CLOUD'23 conference. Congratulations!!
2023/5/4: Somin's CCGrid poster won the best research poster award at CCGrid'23 conference. Congratulations!!!
2023/2/18: Our paper, "QaaD (Query-as-a-Data): Scalable Execution of Massive Number of Small Queries in Spark" is accepted to SIGMOD'23 conference!!!
2023/2/18: Two poster papers by Somin and Fatima accepted to CCGrid'23 conference!
2023/2/3: 김유양, ERC워크샵에서 최우수 포스터 발표상 수상. 권은구, 우수 포스터 발표상 수상! Congratulations!!
2022/12/26: Undergraduate intern, Minyeong Kim (김민영), joined the lab. Welcome!
2022/12/24: Our paper, "A Close Look At SharedResourceConsumption in NoSQL Databases for Accurate Accounting" is accepted to NOMS'23 conference!!
2022/12/14: 석사 졸업생 이재륜 카카오뱅크 합격! Jaeryun Lee received an offer from Kakao Bank. Congratulations!!
2022/12/7: 송소민 KSC 논문 우수상 수상. Somin's paper received the outstanding paper award at KSC. Congratulations!!
2022/12/5: 송소민, 미국 CISCO 본사 인턴 합격! Somin received an internship offer from CISCO, USA. Congratulations!!
2022/9/24: Our paper is accepted to ICCE-Asia conference. Congratulations!!
2022/7/15: 김유양, 송소민. ERC워크샵에서 최우수 포스터 발표상 수상. Congratulations!!
2022/7/04: Undergraduate interns, 권은구, 민시온 joined the lab. Welcome!
2022/6/30: 석사과정 이재륜, 송소민 대학원 우수논문상 입상. Jaeryun and Somin are awarded the Thesis Award of CSE dept. Congratulations!!
2022/6/29: 석사 졸업생 채윤창 삼성전자 합격! Yunchang Chae passed final intervew at Samsung Electronics. Contratulations!!
2022/5/20: 석사과정 장선우 AWS 합격! Sunwoo Jang received an offer from AWS, Korea. Congratulations!!!
2022/4/21: Our paper, "Workload-driven Analysis on the Performance Characteristics of GPU-accelerated DBMSes" is accepted to IEICE Transactions on Information Systems!!
2022/3/31: Our paper, "SecQuant: Quantifying Container System Call Exposure" is accepted to ESORICS'22 conference!!
2022/2/23: 석사 졸업생 서창호 카카오 합격! Changho Seo received an offer from Kakao. Congratulations!!!
2022/2/11: Youyang and Somin received Poster presentation award at ERC workshop, Jeju. Congratulations!
2021/12/22: Our paper, "A Comprehensive Empirical Study of Query Performance Across GPU DBMSes" is accepted to SIGMETRICS'22 conference!!
2021/12/17: Our paper, "NoSQL Database Performance Diagnosis through System Call-level Introspection" is accepted to NOMS'22 conference!!
2021/12/13: "고품질 빅데이터 분석을 위한 최적의 전처리 순열 추천 방법"이 2021년도 한국정보과학회 소사이어티우수논문상 수상!
2021/10/29: Our paper, "A Survey of IoT Stream Query Execution Latency Optimization within Edge and Cloud" is accepted to Wireless Communications and Mobile Computing (WCMC) journal.
2021/10/19: Our paper, "Lognroll: Discovering Accurate Log Templates by Iterative Filtering" is accepted to Middleware'21 conference!
2021/9/1: Master student, Kim Youyang, has joined the lab. Welcome!
2021/6/10: Our paper, "Block-Level Storage Caching for Hypervisor-Based Cloud Nodes" is accepted to IEEE Access journal.
2021/6/1: ERC(Engineering Research Center, $14M USD, 7 years) proposal accepted!!!
2020/12/20: Our paper, "Privacy-aware collaborative task offloading in fog computing" is accepted to IEEE Transactions on Computational Social Systems (TCSS).
2020/9/1: Master student, Song Somin and Jang Sunwoo, have joined the lab. Welcome!
2019/12/9: Our paper, "Green Fog Planning for Optimal Internet-of-Thing Task Scheduling" is accepted to IEEE Access journal.
2019/11/24: Our paper, "BlackEye: Automatic IP Blacklisting using Machine Learning from Security Logs" is accepted to Wireless Networks journal.
2019/11/6: Our paper, "Priolog: mining important logs via temporal analysis and prioritization" is accepted to Sustainability journal. This work is a collaboration with IBM TJ Watson Research Center.
2019/5/30: Our paper, "Measuring the Importance of Log using Temporal Correlation " is accepted to BIC 2019.
2019/5/1: Our paper, "IP Blacklisting via Threat Modeling and Machine Learning from Security Logs " is accepted to ISAIR2019.
2018/12/14: Our paper, "LADRA: Log-based Abnormal Task Detection and Root-cause Analysis in Big Data Processing with Spark " is accepted for publication to the Future Generation Computer Systems (FGCS) journal.
2018/11/2: 서창호 컴퓨터 학부 영예장학생 선발
2018/10/28: Our paper, "Accurate and Efficient Log Template Discovery Technique" is published to KSCI(한국컴퓨터정보학회), 2018.
2018/7/27: 한웅제,서창호 발표 논문이 KCC2018 학부생/주니어논문경진대회 우수상 입상
2018/6: International Conference on Web Services (ICWS) Best paper award!
2018/6: 학부연구생 한웅제, 서창호 - 제주도에서 열리는 KCC 참석 및 논문 발표
2018/5: 차세대컴퓨팅학회 발표 논문, "컨테이너 이미지의 보안 취약성 데이터 수집 및 분석", 우수상 수상
2018/5: Our paper, "컨테이너 이미지의 보안 취약성 데이터 수집 및 분석(Vulnerability Data Collection and Security Analysis of Container Images)" is accepted to 차세대컴퓨팅학회 2018.
2018/5: Our paper, "로그 분석을 위한 로그 템플릿 추출 기법(Log Template Extraction Technique for Log Analysis)" is accepted to 차세대컴퓨팅학회 2018.
2018/5: Our paper, "라이브러리 인트로스펙션을 통한 NoSQL 데이터베이스 동작의 비교 분석" is accepted to KCC 2018.
2018/4/21: Our paper, "EDGESTORE: A Single Namespace and Resource-aware Federation File System for Edge Servers" is accepted to IEEE EDGE 2018 conference.
2017/4/26: Our paper, "Understanding Security Implications of Using Containers in the Cloud" is accepted to USENIX ATC 2017 conference.
2017/3/10: Our paper, "Log-based Abnormal Task Detection and Root Cause Analysis for Spark" is accepted to IEEE ICWS 2017 conference.