Abstract: Scalable Distributed Data Structures (SDDS) consists of two components dynamically spread across a multicomputer: records belonging to a file and a mechanism controlling record placement in file space. Record placement in the file is SDDS specific mechanism. It is spread between SDDS servers, their clients and dedicated process called split coordinator. In the paper fault-tolerant protocols for SDDS components are given. The protocols use Job Comparison Technique along with TMR. Basic and extended SDDS architectures are compared with the help of SDDS oriented software fault injector. Time overhead due to redundancy introduced is estimated, too.
B I B L I O G R A F I A1. Litwin, W., Neimat, M.-A., Schneider, D.: LH*: A Scalable Distributed Data Structure. ACM Transactions on Database Systems ACM-TODS (December 1996)
2. Litwin, W., Neimat, M.-A.: High-Availability LH* Schemes with Mirroring. In: Intl. Conf. on Coope. Inf. Syst. COOPIS-96, Brussels (1996)
3. Litwin, W., Risch, T.: LH*g: a High-availability Scalable Distributed Data Structure by record grouping. U-Paris 9 Tech. Rep. (May 1997)
4. Litwin, W., Menon, J., Risch, T.: LH* Schemes with Scalable Availability. IBM Almaden Research Rep. (May 1998)
5. Litwin, W., Schwarz, T.: LH*RS: A High-Availability Scalable Distributed Data Structure using Reed Solomon Codes. In: CERIA Res. Rep. & ACM-SIGMOD (2000)
6. Sapiecha, K., Lukawski, G.: Fault-tolerant Control for Scalable Distributed Data Structures. Annales Universitatis Mariae Curie-Sklodowska, Informatica (2005)
7. Pradhan, D.K.: Fault-Tolerant Computing: Theory and Techniques. Prentice-Hall, Inc., Englewood Cliffs (1986)
8. Basic Spin Manual: http://spinroot.com/spin/Man/Manual.html 
Pełny tekst 
DOI 
Web of Science