Mitsubishi Electric Unveils High-Performance Sensor Database

Japan’s Mitsubishi Electric Corp. has recently developed a high-performance sensor database that performs quick storage, searches and aggregation of huge data collected from sensors that are anticipated to be used widely in the Internet of Things (IoT) era.

Through the database, sensor data can be used quickly for purposes such as maintaining railways, roads and other infrastructure, tracking factories and managing energy use in homes and buildings. It can process up to 100 trillion data items, such as three-dimensional measurements by laser sensors on/around roads totaling 200,000 kilometers in length, or sensor data collected in a factory over a three-year period using 100,000 sensors that take measurements every 100 milliseconds.

Compared to existing databases, such as commonly used relational databases, Mitsubishi Electric has proven that its high-performance sensor database minimizes load time, storage space, and search and aggregation time each to an array of just one-tenth to one-thousandth of current levels. This was attained by enhancing:

• data compression, to reduce input/output data traffic by choosing compression patterns from over 700 combinations
• data processing units, by processing data in cache memory where possible to improve parallel processing

• data location, by arranging compressed data in storage blocks to reduce input/output time

Huge data-processing performance can be optimized through hardware enhancement, such as distributed or parallel processing utilizing a number of servers, fast storage devices using flash memory, or in-memory processing using huge amounts of memory. However, all of these methods need extremely expensive hardware. Notably, Mitsubishi Electric’s high-performance sensor database attains quick processing using a single server with only one or two CPUs and 4GB of main memory.

Usually, the amount of sensor data in a database increases with long-term use. Some trial systems begin with small amounts of data for verification before full-scale operation utilizing larger data. The high-performance sensor database of Mitsubishi Electric enables incremental expansion of servers as the amount of data grows. Performance and storage capacity are improved in the absence of changes or data migration in application software.