Loki Overview

 

Validation of distributed systems using fault injection is difficult because of their inherent complexity, lack of a global clock, and lack of an easily accessible notion of a global state. To address these challenges, the Loki fault injector injects faults based on a partial view of the global state of the distributed system, and performs a post-runtime analysis using an off-line clock synchronization algorithm to determine whether the faults were properly injected.

The Loki runtime allows dynamic entry and exit of nodes in the system. It also offers an efficient multicast of notification messages and an efficient communication between state machines on the same host.

A new and flexible method for obtaining a wide range of performance and dependability measures also exists in Loki. These measures are based on a flexible language using predicates, and statistical estimations for the results from the correct fault injections.

 

For more information: http://www.crhc.uiuc.edu/PERFORM/loki.html

 

Publications

Conference Publications

            M. Cukier, R. Chandra, D. Henke, J. Pistole, and W. H. Sanders, Fault Injection Based on the Partial Global State of a Distributed System, in Proc. 18th IEEE Symposium on Reliable Distributed Systems (SRDS-99), (Lausanne, Switzerland), IEEE Computer Society Press, pp. 168-177, October 1999.

            R. Chandra, R. M. Lefever, M. Cukier, and W. H. Sanders, Loki: A State-Driven Fault Injector for Distributed Systems, in Proc. International Conference on Dependable Systems and Networks (FTCS-30 and DCCA-8), (New York City, New York, USA), pp. 237-242, June 2000.

            R. Chandra, M. Cukier, R. M. Lefever, and W. H. Sanders, Dynamic Node Management and Measure Estimation in a State-Driven Fault Injector, in Proc. 19th IEEE Symposium on Reliable Distributed Systems (SRDS-2000), (Nürnberg, Germany), pp. 248-257, October 2000.

            K. R. Joshi, M. Cukier, and W. H. Sanders, Experimental Evaluation of the Unavailability induced by a Group Membership Protocol, in Proc. 4th European Dependable Computing Conference (EDCC-4), Toulouse, France, October 23-25, 2002, to appear.

Submitted Journal Publications

            R. Chandra, R. M. Lefever, K. Joshi, M. Cukier, and W. H. Sanders, A Global-State-Triggered Fault Injector for Distributed System Evaluation.