W
Walter B. Ligon
Researcher at Clemson University
Publications - 43
Citations - 1595
Walter B. Ligon is an academic researcher from Clemson University. The author has contributed to research in topics: File system & Reconfigurable computing. The author has an hindex of 15, co-authored 43 publications receiving 1566 citations.
Papers
More filters
PVFS: a parallel file system for linux clusters
TL;DR: The design and implementation of PVFS are described and performance results on the Chiba City cluster at Argonne are presented, both for a concurrent read/write workload and for the BTIO benchmark.
Proceedings ArticleDOI
Implementation and performance of a parallel file system for high performance distributed applications
Walter B. Ligon,Robert Ross +1 more
TL;DR: The Parallel Virtual File System (PVFS) is presented, a system that provides disk striping across multiple nodes in a distributed parallel computer and file partitioning among tasks in a parallel program and which achieves excellent speedups in accessing moderately sized file segments.
Journal ArticleDOI
Characterization of Bandwidth-Aware Meta-Schedulers for Co-Allocating Jobs Across Multiple Clusters
TL;DR: A bandwidth-centric job communication model that captures the interaction and impact of simultaneously co-allocating jobs across multiple clusters is presented and the performance of multi-cluster scheduling algorithms that focus not only on node resource allocation, but also on shared inter-clusters network bandwidth is evaluated.
Journal ArticleDOI
An Exploration of Designing a Hybrid Scale-Up/Out Hadoop Architecture Based on Performance Measurements
TL;DR: A thorough performance measurement of different applications on scale-up and scale-out clusters, configured with Hadoop Distributed File System (HDFS) and a remote file system (i.e., OFS), finds that using OFS rather than HDFS can solve the data storage challenge.
Proceedings ArticleDOI
BMI: a network abstraction layer for parallel I/O
TL;DR: The buffered message interface (BMI) has low processor overhead, minimal impact on latency, and can improve throughput for parallel file system workloads by as much as 40% compared to other more generic network abstractions.