Title :
RAFlow: Read Ahead Accelerated I/O Flow through Multiple Virtual Layers
Author :
Zhaoning Zhang ; Kui Wu ; Huiba Li ; Jinghua Feng ; Yuxing Peng ; Xicheng Lu
Author_Institution :
Nat. Key Lab. for Parallel & Distrib. Process., Nat. Univ. of Defense Technol., Changsha, China
Abstract :
Virtualization is the foundation for cloud computing, and the virtualization can not be achieved without software defined, elastic, flexible and scalable virtual layers. Unfortunately, if multiple virtual storage devices are chained together, the system may be subject to severe performance degradation. While the read-ahead (RA) mechanism in storage devices plays a very important role to improve I/O performance, RA may not be effective as expected for multiple virtualization layers, since it is originally designed for one layer only. When I/O requests are passed through a long I/O path, they may trigger a chain reaction and lead to unnecessary data transmission and thus bandwidth waste. In this paper, we study the dynamic behavior of RA through multiple I/O layers and demonstrate that if controlled well, RA can greatly accelerate I/O speed. We present RAFlow, a RA control mechanism, to effectively improve I/O performance by strategically expanding RA window at each layer. Our real-world experiments show that it can achieve 20% to 50% performance improvement in I/O paths with up to 8 virtualized storage devices.
Keywords :
cloud computing; input-output programs; virtualisation; I/O path; I/O performance; I/O requests; RA mechanism; RAFlow; cloud computing; input-output flow; read ahead accelerated I/O flow; read-ahead mechanism; virtual storage devices; virtualization; Acceleration; Benchmark testing; Degradation; Distributed databases; Performance evaluation; Radio access networks; Synchronization; Cloud Computing; I/O Performance; IaaS Storage; Read ahead;
Conference_Titel :
Networking, Architecture, and Storage (NAS), 2014 9th IEEE International Conference on
Conference_Location :
Tianjin
DOI :
10.1109/NAS.2014.13