Speeding up CGM-Based Parallel Algorithm for Minimum Cost Parenthesizing Problem
Year: 2018
Type: conference-paper
Abstract: Parallelization of dynamic programming algorithms is a very well studied topic. This paper presents a parallel algorithm on the Coarse Grained Multicomputer model (CGM for short) for solving the Optimal String Parenthesizing Problem or Minimum Cost Parenthesizing Problem. In the previous best CGM algorithm for this problem, which uses regular partitioning of the dynamic graph, it is difficult to simultaneously optimize load-balancing of the processors and minimization of the communication time in the same algorithm. In this paper, we try to bring a solution to these two contradictory objectives. We propose a new technique of irregular partitioning of this graph which allows processors to stay active as long as possible. This promotes load-balancing and thus minimizes the overall computation time of processors. This also reduces their latency, which is the largest part of the overall communication time.
Source: HAL
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4289740682