Guillermo A. Alvarez

Declustered disk array architectures with optimal and near-optimal parallelism

This paper investigates the placement of data and parity on redundant disk arrays. Declustered organizations have been traditionally used to achieve fast reconstruction of a failed disks contents. In previous work, Holland and Gibson identified six desirable properties for ideal layouts; however, no declustered layout satisfying all properties has been published in the literature. We present a complete, constructive characterization of the collection of ideal declustered layouts possessing all six properties. Given that ideal layouts exist only for a limited set of configurations, we also present two novel layout families. PRIME and RELPR can tolerate multiple failures in a wide variety of configurations with slight deviations from the ideal. Our simulation studies show that the new layouts provide excellent parallel access performance and reduced incremental loads during degraded operation, when compared with previously published layouts. For large accesses and under high loads, response times for the new layouts are typically smaller than those of previously published declustered layouts by a factor of 2.5.

High-performance asynchronous atomic broadcast

We describe two families of asynchronous atomic broadcast protocols that provide good delivery and stability times, use a small number of messages to accomplish a broadcast, distribute the load of ordering messages evenly among group members, use efficient flow-control techniques, and provide gracefully degraded performance in the presence of communication failures. The pinwheel protocols are designed for applications characterized by a uniform message arrival pattern. The on-demand protocol is designed for non-uniform, bursty message arrivals. The protocols tolerate omission/performance communication failures and crash/performance process failures. Simulation studies for the pinwheel protocols demonstrate that they have superior performance over other well known atomic broadcast protocols, for uniform message arrival rates. We also report measurements taken on prototype implementations, that show very good results and a substantial performance advantage of the on-demand protocol for non-uniform and intermediate message arrival patterns.

Applying simulation to the design and performance evaluation of fault-tolerant systems

The paper illustrates how the CESIUM simulation tool can be used for design and performance evaluation of fault tolerant and real time systems, in addition to testing the correctness of protocol implementations. We calibrate three increasingly accurate simulation models of a network of workstations using independently obtained data. For a sample group membership protocol, the predictions of the simulator are very close to the actual performance measured in the real system. We also apply CESIUM to the evaluation of two potential improvements for the protocol, performing experiments that would have been difficult to implement in the real system. The results of the simulations give us valuable insight on how to tune configuration parameters, as well as on the performance gains of the improved versions. Our experience shows that CESIUM can be used to develop best effort services which adapt their quality of service according to the failures that occur during operation.

Centralized failure injection for distributed, fault-tolerant protocol testing

We describe a centralized approach to testing that distributed fault-tolerant protocols satisfy their safety and timeliness specifications in the presence of the very failures they are designed to tolerate. CESIUM is a testing environment based on the centralized simulation of distributed executions and failures. Processes are run in a single address space while providing the appearance of a truly distributed execution. The human tester can force the occurrence of arbitrary failures and security attacks. The implementations under test are not instrumented for testing purposes, and their source codes need not be available. We prove that CESIUM can execute exactly the set of runs feasible in the real distributed system being simulated. We also show that there are safety and timeliness properties in the specifications of many existing distributed protocols that cannot be tested in practical distributed systems. All of these properties can, however, be accurately tested by CESIUM without introducing any perturbation in test experiments.

Run-time support for asynchronous parallel computations

In this peeper we describe DREAM, a distributed environment that provides run-time support for parallel computations on asynchronous multiprocessors. The system supports global distributed arrays as collections of subarrays in the local memories of the intervening processors. Nodes allocate and deallocate array portions dynamically, and access external array sections without the intervention of the user programming in the remote node. Remote accesses can be performed while the program continues its execution, thus overlapping communication and computation. This feature allows the user to implement dynamic communication patterns by accessing external array elements on demand without incurring a heavy performance penalty. DREAM also vectorizes requests into larger network messages for efficiency. We report performance results for an application running on top of a prototype of the system, showing good scalability and masking the network latency with computation.<<ETX>>

Efficient management of multiple outstanding timeouts

Abstract The timer management service maintains a set of independent timeout intervals. Client applications initiate and terminate timers, and are notified when timeouts occur. We present and prove the correctness of a scheme that implements this service, and compares favorably with previously existing solutions.