Bridging the Processor/Memory Performance Gap
in Database Applications

Anatassia Ailamaki
Carnegie Mellon University

Friday, December 14, 2001, 4:00 PM
Volen 106, Brandeis University

Abstract:

Database applications are predominantly memory-intensive workloads, and their performance is substantially influenced by the memory access latency. Memory speeds, however, have fundamentally lagged behind processor speeds. Today's memory systems incur access latencies that are up to three orders of magnitude larger than the latency of a single arithmetic operation. Previous work has demonstrated that database system performance suffers from memory access delays incurred by data and instruction cache misses, and similar performance trends exist across commercial database systems. Therefore, to significantly improve database system behavior on modern processor platforms database software designers should focus on maximizing cache utilization and keep data that are likely to be referenced in the cache hierarchy.

This talk will analyze the impact of data placement on database system performance. The data placement scheme used in today's database systems "pushes" unreferenced data to caches, wasting memory bandwidth, polluting the cache, and exposing hard-to-overlap memory latencies. I will introduce a novel data placement scheme, called Partition Attributes Across (PAX). PAX eliminates unnecessary memory accesses by only bringing useful data into the cache. The experimental results on a variety of workloads show that PAX drastically reduces data-related stalls and therefore elapsed execution time drops to half. Time permitting, I will also present an overview of the dynamic cache-aware database placement techniques I am currently working on, as well as preliminary results.

Speaker Bio:

Anastassia Ailamaki received a B.Sc. degree in Computer Engineering (1990) from the Polytechnic School of the University of Patra, Greece, M.Sc. degrees from the Technical University of Crete, Greece and from the University of Rochester, NY, and a Ph.D. degree in Computer Science from the University of Wisconsin-Madison. She is currently an assistant professor at the Carnegie Mellon School of Computer Science. Her recent work on cache-conscious data placement received a best-paper award in VLDB 2001. Her current research interests include database system design and performance, cache-resident databases, internet querying and caching, workload characterization, and scientific workflow management systems.

Maintained by Dina Goldin dqg AT cse.uconn.edu
Last updated on 12/4/01