Micron Releases HSE 3.0 Open-Source Storage Engine

In early 2020 the software engineers at Micron announced an open-source storage engine designed for SSDs and persistent memory. The storage engine prided itself on being a speedy key-value store database and there was also a MongoDB-based implementation. Last year HSE 2.0 debuted and the updated storage engine no longer relied on modifications to the Linux kernel and now strictly a user-space based solution. This week the Micron engineers are celebrating HSE 3.0 as their latest work on this open-source storage engine.

As a refresher, the HSE project describes itself as:

"HSE is a fast embeddable key-value store designed for SSDs and persistent memory. HSE optimizes performance and endurance by orchestrating data placement across DRAM and multiple classes of solid-state storage.

HSE is ideal for powering Databases, Software-Defined Storage (SDS), High-Performance Computing (HPC), Internet of Things (IoT), and Machine Learning (ML)."

HSE 3.0 was released on Thursday and has improved data management for better performance across various important workloads. The HSE 3.0 engine has made a variety of different performance optimizations around workloads with monotonically increasing keys / time-series data, "many client workloads", the ability to store compressed and uncompressed values together, and other improvements in the name of performance.

Java language bindings are also now available for HSE 3.0 and complement the updated Python bindings.

For downloads or to learn more about this SSD-optimized open-source storage engine see the HSE project on GitHub.