Proportional Service Allocation in Distributed Storage Systems ∗

Fully distributed storage systems have gained popularity i n he past few years because of their ability to use cheap commodity hardware and their high scalability. While there are a number of algorithms for providing dif ferentiated quality of service to clients of a centralized storage system, the prob lem has not been solved for distributed storage systems. Providing performance gu arantees in distributed storage systems is more complex because clients may have dif ferent data layouts and access their data through different coordinators (acce ss nodes), yet the performance guarantees required are global. This paper presents a distributed scheduling framework. It is an adaptation of fair queuing algorithms for distributed servers. Specifica lly, upon scheduling each request, it enforces an extra delay (possibly zero) that cor responds to the amount of service the client gets on other servers. Different perfo rmance guarantees, e.g., per storage node proportional sharing, total service propo rtional sharing or mixed, can be met by different delay functions. The delay functions can be calculated at coordinators locally so excess communication is avoided. T he analysis and experimental results show that these new algorithms can enfor ce performance goals under different data layouts and workloads.

[1]  Anand Sivasubramaniam,et al.  An interposed 2-Level I/O scheduling framework for performance virtualization , 2005, SIGMETRICS '05.

[2]  Gang Peng,et al.  Multi-dimensional storage virtualization , 2004, SIGMETRICS '04/Performance '04.

[3]  Hui Zhang,et al.  Service disciplines for guaranteed performance service in packet-switching networks , 1995, Proc. IEEE.

[4]  Upamanyu Madhow,et al.  Fair scheduling with tunable latency: a round-robin approach , 2003, TNET.

[5]  Scott Shenker,et al.  Analysis and simulation of a fair queueing algorithm , 1989, SIGCOMM '89.

[6]  Xiaoyun Zhu,et al.  Triage: performance isolation and differentiation for storage systems , 2004, Twelfth IEEE International Workshop on Quality of Service, 2004. IWQOS 2004..

[7]  Chandramohan A. Thekkath,et al.  Petal: distributed virtual disks , 1996, ASPLOS VII.

[8]  Arif Merchant,et al.  Façade: Virtual Storage Devices with Performance Guarantees , 2003, FAST.

[9]  Haiyun Luo,et al.  A Packet Scheduling Approach to QoS Support in Multihop Wireless Networks , 2004, Mob. Networks Appl..

[10]  Banu Özden,et al.  Disk scheduling with quality of service guarantees , 1999, Proceedings IEEE International Conference on Multimedia Computing and Systems.

[11]  Hui Zhang,et al.  Implementing distributed packet fair queueing in a scalable switch architecture , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[12]  Christos Faloutsos,et al.  Storage device performance prediction with CART models , 2004, The IEEE Computer Society's 12th Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, 2004. (MASCOTS 2004). Proceedings..

[13]  Paramvir Bahl,et al.  Distributed fair scheduling in a wireless LAN , 2000, IEEE Transactions on Mobile Computing.

[14]  Wei Jin,et al.  Interposed proportional sharing for a storage service utility , 2004, SIGMETRICS '04/Performance '04.

[15]  Edward W. Knightly,et al.  Distributed Priority Scheduling and Medium Access in Ad Hoc Networks , 2002, Wirel. Networks.

[16]  Arif Merchant,et al.  FAB: building distributed enterprise disk arrays from commodity components , 2004, ASPLOS XI.

[17]  Jian Xu,et al.  Performance virtualization for large-scale storage systems , 2003, 22nd International Symposium on Reliable Distributed Systems, 2003. Proceedings..

[18]  Arif Merchant,et al.  A modular, analytical throughput model for modern disk arrays , 2001, MASCOTS 2001, Proceedings Ninth International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[19]  Scott Shenker,et al.  Analysis and simulation of a fair queueing algorithm , 1989, SIGCOMM 1989.

[20]  Abhay Parekh,et al.  A generalized processor sharing approach to flow control in integrated services networks-the single node case , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.

[21]  Peter J. Varman,et al.  Lexicographic QoS scheduling for parallel I/O , 2005, SPAA '05.

[22]  Jeffrey Katcher,et al.  PostMark: A New File System Benchmark , 1997 .

[23]  Yixin Diao,et al.  Feedback Control of Computing Systems , 2004 .

[24]  Lixia Zhang.  VirtualClock: A New Traffic Control Algorithm for Packet-Switched Networks , 1991, ACM Trans. Comput. Syst..

[25]  Arif Merchant,et al.  A decentralized algorithm for erasure-coded virtual disks , 2004, International Conference on Dependable Systems and Networks, 2004.

[26]  H. Vin,et al.  Start-time fair queueing: a scheduling algorithm for integrated services packet switching networks , 1996, SIGCOMM '96.

[27]  Zoran Dimitrijevic,et al.  Quality of Service Support for Real-time Storage Systems , 2003 .

[28]  Carl A. Waldspurger,et al.  Memory resource management in VMware ESX server , 2002, OSDI '02.