Year: 2017
Type: conference-paper
Abstract: We consider the problem of computing the convolution of two long vectors using parallel processors in the presence of “stragglers”. Stragglers refer to the small fraction of faulty or slow processors that delays the entire computation in time-critical distributed systems. We first show that splitting the vectors into smaller pieces and using a linear code to encode these pieces provides improved resilience against stragglers than replication-based schemes under a simple, worst-case straggler analysis. We then demonstrate that under commonly used models of computation time, coding can dramatically improve the probability of finishing the computation within a target “deadline” time. As opposed to the more commonly used technique of expected computation time analysis, we quantify the exponents of the probability of failure in the limit of large deadlines. Our exponent metric captures the probability of failing to finish before a specified deadline time, i.e., the behavior of the “tail”. Moreover, our technique also allows for simple closed form expressions for more general models of computation time, e.g. shifted Weibull models instead of only shifted exponentials. Thus, through this problem of coded convolution, we establish the utility of a novel asymptotic failure exponent analysis for distributed systems.
Institutions: Carnegie Mellon University, Pennsylvania State University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: CRII: CIF: Towards a Systematic Interference Alignment Approach for Network Information Flow, CAREER: An Information Theoretic Perspective of Consistent Distributed Storage Systems, CAREER: Towards Green Communications Using an Information-Lens: Foundations of the Joint Design of Communication Strategies and Circuits
OpenAlex ID: W2614091483
