Measuring the cost of scalability and reliability for Internet-based, server-centered applications

With large numbers of geographically dispersed clients, a centralized approach to Internet-based application development is not scalable and also not dependable. This paper presents a decentralized approach to dependable Internet-based application development, consisting of a logical structuring of collaborating sub-systems of geographically separated replicated servers. Two implementations of an Internet auction, one using a centralized approach and the other using our decentralized approach, are described. To evaluate the scalability of the two approaches, a number of experiments are performed on these implementations and the results are presented.

[1]  Mark Garland Hayden,et al.  The Ensemble System , 1998 .

[2]  Israel Ben-Shaul,et al.  GEM: A Global Electronic Market System , 1999, Inf. Syst..

[3]  Sanjoy Paul,et al.  Reliable Multicast Transport Protocol (RMTP) , 1997, IEEE J. Sel. Areas Commun..

[4]  Kenneth P. Birman,et al.  The process group approach to reliable distributed computing , 1992, CACM.

[5]  P. Klemperer.  Auction Theory: A Guide to the Literature , 1999 .

[6]  Nancy A. Lynch,et al.  Impossibility of distributed consensus with one faulty process , 1983, PODS '83.

[7]  Paul D. Ezhilchelvan,et al.  Design and implemantation of a CORBA fault-tolerant object group service , 1999, DAIS.

[8]  Louise E. Moser,et al.  Totem: a fault-tolerant multicast group communication system , 1996, CACM.

[9]  Kenneth P. Birman.  The Process Group Approach to Reliable , 2000 .

[10]  Arne Andersson,et al.  Managing large scale computational markets , 1998, Proceedings of the Thirty-First Hawaii International Conference on System Sciences.

[11]  Paul D. Ezhilchelvan,et al.  Principal Features of the VOLTAN Family of Reliable Node Architectures for Distributed Systems , 1992, IEEE Trans. Computers.

[12]  Newtop: a fault-tolerant group communication protocol , 1995, Proceedings of 15th International Conference on Distributed Computing Systems.

[13]  Yair Amir,et al.  Transis: a communication subsystem for high availability , 1992, [1992] Digest of Papers. FTCS-22: The Twenty-Second International Symposium on Fault-Tolerant Computing.

[14]  Miguel Oom Temudo de Castro,et al.  Practical Byzantine fault tolerance , 1999, OSDI '99.

[15]  Paul D. Ezhilchelvan,et al.  An end-to-end reliable multicast protocol using polling for scaleability , 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.

[16]  Sam Toueg,et al.  Fault-tolerant broadcasts and related problems , 1993 .

[17]  Rogério de Lemos,et al.  A robust group membership algorithm for distributed real-time systems , 1990, [1990] Proceedings 11th Real-Time Systems Symposium.

[18]  Rachid Guerraoui,et al.  The Implementation of a CORBA Object Group Service , 1998, Theory Pract. Object Syst..

[19]  F. Cristian.  Reaching Agreement on Processor Group Membership in Synchronous Distributed Systems Key Words: Communication Network { Distributed System { Failure Detection { Fault Tolerance { Real Time System { Replicated Data , 1991 .

[20]  Yair Amir,et al.  Transis: A Communication Sub-system for High Availability , 1992 .

[21]  Madhu Sudan,et al.  A reliable dissemination protocol for interactive collaborative applications , 1995, MULTIMEDIA '95.

[22]  Stephen Deering,et al.  Multicast routing in a datagram internetwork , 1992 .