Benchmarking SciDB Data Import on HPC Systems

TL;DR

This paper evaluates SciDB's performance for large-scale imaging data management on HPC systems, demonstrating high insert rates and efficient random data access using D4M and supercomputing techniques.

Contribution

It presents the first performance benchmarking of SciDB on HPC systems with simulated imaging data, utilizing D4M and parallel techniques.

Findings

Peak insert performance of 2.2 million per second on a single node

Efficient random sub-volume access surpassing traditional file-based methods

Effective use of parallel inserts and distributed arrays

Abstract

SciDB is a scalable, computational database management system that uses an array model for data storage. The array data model of SciDB makes it ideally suited for storing and managing large amounts of imaging data. SciDB is designed to support advanced analytics in database, thus reducing the need for extracting data for analysis. It is designed to be massively parallel and can run on commodity hardware in a high performance computing (HPC) environment. In this paper, we present the performance of SciDB using simulated image data. The Dynamic Distributed Dimensional Data Model (D4M) software is used to implement the benchmark on a cluster running the MIT SuperCloud software stack. A peak performance of 2.2M database inserts per second was achieved on a single node of this system. We also show that SciDB and the D4M toolbox provide more efficient ways to access random sub-volumes of massive datasets compared to the traditional approaches of reading volumetric data from individual files. This work describes the D4M and SciDB tools we developed and presents the initial performance results. This performance was achieved by using parallel inserts, a in-database merging of arrays as well as supercomputing techniques, such as distributed arrays and single-program-multiple-data programming.