HFetch: Hierarchical Data Prefetching for Scientific Workflows in Multi-Tiered Storage Environments
Authors: H. Devarajan, A. Kougkas, X.-H. Sun
Date: May, 2020
Venue: IEEE International Parallel and Distributed Processing Symposium (IPDPS'20), May 18-22, 2020
Type: Conference
Abstract
In the era of data-intensive computing, accessing data with a high-throughput and low-latency is more imperative than ever. Data prefetching is a well-known technique for hiding read latency. However, existing solutions do not consider the new deep memory and storage hierarchy and also suffer from under-utilization of prefetching resources and unnecessary evictions. Additionally, existing approaches implement a client- pull model where understanding the application's I/O behavior drives prefetching decisions. Moving towards exascale, where machines run multiple applications concurrently by accessing files in a workflow, a more data-centric approach can resolve challenges such as cache pollution and redundancy. In this study, we present HFetch, a truly hierarchical data prefetcher that adopts a server-push approach to data prefetching. We demonstrate the benefits of such an approach. Results show 10- 35% performance gains over existing prefetchers and over 50% when compared to systems with no prefetching.