数据中心环境下针对突发性io的调度研究-research on scheduling of bursty io in data center environment.docxVIP

华中科技大学硕士学位论文 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract With the increasing popularity of the Internet, the data volume has grown rapidly in both individuals and business area. Thus how to store such large amounts of data in a convenient and safe manner becomes a hot research issue in the field of storage. A data center characterized by storage space rental and shared storage facilities is a feasible solution for data storage in cloud era. A new technical challenge in the data center is to better manage storage resources to provide a high quality of service (QoS), which usually sparks the interest of many researchers. Burst I/O is a common phenomenon within storage workloads, i.e. unexpected intensive I/O flow in a random period of time. If we deploy a data center according to the highest demand, it will increase both storage rationing and management cost; thus such burst I/O phenomenon results in the increase of the storage service requirement. We introduce a workload scheduling framework to solve the burst I/O problem, where the burst I/Os are dynamically decomposed and isolated and the decomposed I/Os are then restructured. First, we survey the related researches on workload scheduling framework, and analyze the impact of burst I/O on performance, and summarize the approaches to resources assignment used in the data centers. Second, we design a ‘decomposing + restructuring’ scheduling framework. It is shown that such load scheduling framework can efficiently reduce the cost of server resource distribution under the premise of guaranteed QoS. We design two load-decomposing schemes and a load-restructuring scheme. All load-scheduling schemes are implemented in a prototype system called Pandaria, and many functional modules are incorporated to the system, including a frame information module, a load scheduling module, a thread pool module, a storage load testing module, and so on. Final, we use DiskSim simulator to assess the performance impact caused by burst I/O, and