On the performance and convergence of distributed stream processing via approximate fault tolerance

Abstract

Fault tolerance is critical for distributed stream processing systems, yet achieving error-free fault tolerance often incurs substantial performance overhead. We present AF-Stream, a distributed stream processing system that addresses the trade-off between performance and accuracy in fault tolerance. AF-Stream builds on a notion called approximate fault tolerance, whose idea is to mitigate backup overhead by adaptively issuing backups, while ensuring that the errors upon failures are bounded with theoretical guarantees. Specifically, AF-Stream allows users to specify bounds on both the state divergence and the loss of non-backup streaming items. It issues state and item backups only when the bounds are reached. Our AF-Stream design provides an extensible programming model for incorporating general streaming algorithms as well as exports only few threshold parameters for configuring approximation fault tolerance. Furthermore, we formally prove that AF-Stream preserves high algorithm-specific accuracy of streaming algorithms, and in particular the convergence guarantees of online learning. Experiments show that AF-Stream maintains high performance (compared to no fault tolerance) and high accuracy after multiple failures (compared to no failures) under various streaming algorithms. * An earlier version of the paper appeared in [39]. In this extended version, we formally prove that AF-Stream preserves the convergence guarantees of online learning in distributed stream processing, and also validate our convergence analysis through prototype experiments for different datasets, parameters, and consistency models.

Optimizing checkpoint‐based fault‐tolerance in distributed stream processing systems: Theory to practice

“…Approximate fault tolerance is another method used to improve fault-tolerance, which addresses the trade-off between accuracy and performance. 33,34 This approach checkpoints the state and the unprocessed data only when the number of errors due to failures is higher than the user-defined bound. Although approximate fault tolerance does not always guarantee accuracy, it gives better performance as the number of performed checkpoints are fewer compared to a system that does periodic checkpoints.…”