클라우드 환경에서 단일 자원 스케줄링 연구
Exploring the limitations of single resource scheduling in clouds
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
클라우드 상에서는 웹 서버, 비디오 스트리밍 등 다양한 응용 프로그램이 동작하며 이러한 응용 프로그램들은 그 특성이 서로 다르고 시간에 따라 변화하는 특징이 있다. 예를 들어, 한 응용 프로그램은 주로 CPU를 소모하는 반면, 다른 응용 프로그램은 네트워크를 많이 사용할 수 있다. 그러나 현재의 클라우드 상에서의 자원 스케줄링은 CPU 스케줄링 또는 네트워크 스케줄링과 같은 단일 자원에만 초점을 맞추고 있다. 본 논문은 다양한 성능 평가 및 프로파일링을 통해 현재의 단일 자원 스케줄링으로 인한 가상 머신의 성능 저하와 그 원인을 분석하여 단일 자원 스케줄링의 한계를 지적한다. 평가 결과, 단일 자원 스케줄링은 평균적으로 최대 성능의 67%와 73%만 제공하는 것으로 나타났다. 평가 결과를 바탕으로, 우리는 단일 자원 스케줄링의 한계를 극복하고 다양한 특성을 갖는 클라우드 응용 프로그램을 지원할 수 있는 새로운 스케줄링 기법의 개발이 필요함을 보인다.
Abstract
Clouds execute very heterogeneous applications such as webservers and video streaming that have different characteristics. For example, one application burns mostly CPU whereas another uses network heavily. Some applications vary their characteristics with time. Yet, resource scheduling in clouds focuses on a single resource -- CPU scheduling or network scheduling. We call it a separate scheduling. Through extensive performance evaluation and profiling, this paper points out the limitations of the separate scheduling, which results in the performance degradation of the time-varying application. Evaluation results show that the separate scheduling only offers 67% and 73% of maximum performance on average. Based on the evaluation results, we argue that it is necessary to develop a new scheduling technique that can overcome the limitation of the separate scheduling and support cloud applications with time-varying characteristics.
참고문헌 (40)
- "Latency is everywhere and it costs you sales"[URL]
- CHENG, L., "Network performance isolation for latency-sensitive cloud applications", Future Generation Computer Systems, 29(4), pp.1073~1084 (2013)
- "5 lessons we’ve learned using aws" (2010)[URL]
- CUCINOTTA, T., "Providing performance guarantees to virtual machines using real-time scheduling", Euro-Par 2010 Parallel... (workshops), pp.657~664 (2010)
- DEAN, J., "The tail at scale", Communications of the ACM, 56(2), pp.74~80 (2013)
- DOGAR, F. R., "Decentralized taskaware scheduling for data center networks", Proceedings of the 2014 ACM conference on SIGCOMM- SIGCOMM ’14, pp.431~442 (2014)
- "Bing and google agree: Slow pages lose users" (2009)[URL]
- GARG, S. K., "SLA-based virtual machine management for heterogeneous workloads in a cloud datacenter", Journal of Network and Computer Applications, 45, pp.108~120 (2014)
- GHODSI, A., "Dominant Resource Fairness: Fair Allocation of Multiple Resource Types Maps Reduces", NSDI, 167, pp.24~24 (2011)
- GUO, C., "Secondnet: a data center network virtualization architecture with bandwidth guarantees", Proceedings of the 6th International Conference, pp.15~ (2010), ACM
- GUPTA, V., "Hetero Visor: Exploiting Resource Heterogeneity to Enhance the Elasticity of Cloud Platforms", Proceedings of the 11th ACM SIGPLAN, 2015, 79~92[DOI]
- "The cost of latency" (2009)[URL]
- JANG, K., "Silo: Predictable Message Latency in the Cloud", Proc. ACM SIGCOMM, pp.435~448 (2015)
- JEYAKUMAR, V., "EyeQ: Practical network performance isolation at the edge", Rem, pp.297~311 (2013)
- KIM, H., "Scheduler Support for Video-oriented Multimedia on Client-side Virtualization", MMSys 2012, pp.65~76 (2012)
- KIM, H., "Virtual asymmetric multiprocessor for interactive performance of consolidated desktops", Proceedings of the 10th ACM SIGPLAN, 2014, 29~40[DOI]
- KIVITY, A., "kvm: the linux virtual machine monitor", Proceedings of the Linux symposium, 1, pp.225~230 (2007)
- LIU, D., "Scheduling paravirtualized virtual machines based on events", Future Generation Computer Systems, 29, pp.1130~1139 (2013)
- NATHUJI, R., "Q-Clouds:Managing Performance Interference Effects for QoS-Aware Clouds", Proceedings of the 5th European conference on Computer systems, pp.237~250 (2010)
- NOVAKOVIC, D., "DeepDive:Transparently Identifying and Managing Performance Interference in Virtualized Environments", Atc ’13 (2013)
- POPA, L., "FairCloud - Sharing the Network in Cloud Computing", Proceedings of the ACM SIGCOMM 2012 conference on Applications, technologies, architectures, and protocols for computer communication -SIGCOMM ’12, 187 (2012)
- PU, X., "Who Is Your Neighbor: Net I", O Performance Interference in Virtualized Clouds, 2013, pp.6(3), 314~329
- RODRIGUES, H., "Gatekeeper: Supporting bandwidth guarantees for multi-tenant datacenter networks", USENIX WIOV (2011)
- SHIEH, A., "Seawall: Performance Isolation for Cloud Datacenter Networks", HotCloud’10, USENIX Association, (3), pp.1~1 (2010)
- SUKWONG, O., "SageShift: Managing SLAs for highly consolidated cloud", Proceedings-IEEE INFOCOM, 208–216 (2012)
- TAN, H., "DMVL: An I", O bandwidth dynamic allocation method for virtual networks, 2014, pp.39, 104~116
- TEMBEY, P., "Merlin: Application- and Platform-aware re-source allocation in consolidated server systems", Proceedings of the ACM Symposium on Cloud Computing - SOCC ’14, (3), pp.1~14 (2014), ACM Press
- VERBOVEN, S., "Network Aware Scheduling for Virtual Machine Workloads with Interference Models", IEEE Transactions on Services Computing, 1374, pp.1~ (2014)
- WANG, G. W. G., "The Impact of Virtualization on Network Performance of Amazon EC2 Data Center", INFOCOM, Proceedings IEEE (2010)
- WANG, H., "A-DRM:Architecture-aware Distributed Resource Management of Virtualized Clusters", Proceedings of the 11th ACM SIGPLAN, 2015, 93~106[DOI]
- XU, C., "vTurbo:Accelerating Virtual Machine I", O Processing Using Designated Turbo-Sliced Core, 2013, 243~254
- "Linux traffic control–implementation overview" (1998)[URL]
- "Quota policy"[URL]
- Chowdhury, M., "Multi-Resource Fairness for Correlated and Elastic Demands", 13th USENIX Symposium on Networked Systems Design and Implementation (NSDI 16), pp.407~424 (2016)
- Koh, Y., "An analysis of performance interference effects in virtual environments", 2007 IEEE International Symposium on Performance Analysis of Systems & Software, pp.200~209 (2007), IEEE
- "Apache HTTP server benchmarking tool"[URL]
- 김진택, "클라우드컴퓨팅 기술 스택 분석", 한국차세대컴퓨팅학회 논문지, 11(6), pp.79~89 (2015), 한국차세대컴퓨팅학회
- 임철수, "클라우드 컴퓨팅 서비스를 위한 SLA기반 다중 임차자 프레임워크 설계", 한국차세대컴퓨팅학회 논문지, 7(4), pp.38~46 (2011), 한국차세대컴퓨팅학회
- 강동화, "KVM 기반 가상화 환경에서 CPU 스케줄링 관점으로 본 Network I", O 성능간섭 현상 분석, 2016, pp.22(9), 473~478[DOI]
- 최동훈, "이질적 고성능 클라우드 컴퓨팅을 위한 확장형 OpenStack의 개발 및 평가", 한국차세대컴퓨팅학회 논문지, 12(3), pp.41~49 (2016), 한국차세대컴퓨팅학회