Active Disk Paxos with infinitely many processes

被引:19
作者
Chockler, G [1 ]
Malkhi, D [1 ]
机构
[1] MIT, Comp Sci & Artificial Intelligence Lab, Cambridge, MA 02139 USA
关键词
shared memory; consensus; Paxos; infinitely many processes; non-responsive object faults;
D O I
10.1007/s00446-005-0123-x
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We present an improvement to the Disk Paxos protocol by Gafni and Lamport which utilizes extended functionality and flexibility provided by Active Disks and supports unmediated concurrent data access by an unlimited number of processes. The solution facilitates coordination by an infinite number of clients using finite shared memory. It is based on a collection of read-modify-write objects with faults, that emulate a new, reliable shared memory abstraction called a ranked register. The required read-modify-write objects are readily available in Active Disks and in Object Storage Device controllers, making our solution suitable for state-of-the-art Storage Area Network (SAN) environments.
引用
收藏
页码:73 / 84
页数:12
相关论文
共 50 条
[1]   Computing with faulty shared objects [J].
Afek, Y ;
Greenberg, DS ;
Merrit, M ;
Taubenfeld, G .
JOURNAL OF THE ACM, 1995, 42 (06) :1231-1274
[2]  
AMIRI K, 2000, P INT C DISTR COMP S
[3]   Serverless network file systems [J].
Anderson, TE ;
Dahlin, MD ;
Neefe, JM ;
Patterson, DA ;
Roselli, DS ;
Wang, RY .
ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1996, 14 (01) :41-79
[4]  
[Anonymous], P 7 INT C ARCH SUPP
[5]  
BIRMAN IK, 1987, P 11 ANN S OP SYST P, P123
[6]  
Boichat R., 2003, SIGACT News, V34, P47, DOI 10.1145/637437.637447
[7]  
BOICHAT R, ID200106 DSC
[8]   BOUNDS ON SHARED-MEMORY FOR MUTUAL EXCLUSION [J].
BURNS, JE ;
LYNCH, NA .
INFORMATION AND COMPUTATION, 1993, 107 (02) :171-184
[9]  
BURNS RC, 2000, THESIS U CALIFORNIA
[10]   Unreliable failure detectors for reliable distributed systems [J].
Chandra, TD ;
Toueg, S .
JOURNAL OF THE ACM, 1996, 43 (02) :225-267