Improving Seek Time for Column Store Using MMH Algorithm

TL;DR

This paper introduces the MMH algorithm, which enhances hash-based search efficiency in column stores by mitigating data distribution effects, leading to significant throughput improvements.

Contribution

The paper proposes the Memory Managed Hash (MMH) algorithm utilizing shift XOR groups to improve hash search performance in column stores.

Findings

MMH improves read throughput by 22% for TPC-H distribution.

MMH enhances write throughput similarly.

The approach reduces the impact of data distribution on hash searches.

Abstract

Hash based search has, proven excellence on large data warehouses stored in column store. Data distribution has significant impact on hash based search. To reduce impact of data distribution, we have proposed Memory Managed Hash (MMH) algorithm that uses shift XOR group for Queries and Transactions in column store. Our experiments show that MMH improves read and write throughput by 22% for TPC-H distribution.