Heterogeneous-Memory Storage Engine

Representing a significant step forward for open source storage engines, HSE 2.0 now makes it even easier to build high-performance storage applications for cloud-native and containerized environments.

Start developing HSE 2.0 at GitHub

HSE from Micron

Introducing HSE, one of the first open source heterogeneous-memory storage engines, designed specifically for solid-state drives and storage-class memory. Legacy storage engines born in the era of hard disk drives failed to architecturally provide for the increased performance and reduced latency of next-generation nonvolatile media.

Watch video

Accelerate your storage applications

Micron’s unique open source HSE was designed to maximize the capabilities of new storage technologies by intelligently and seamlessly managing data among multiple storage classes. The result is significantly improved performance, increased drive endurance and decreased latency — even under the crushing burden of massive-scale deployments. Once again, Micron is moving up the stack to deliver even higher value.

Get started developing at www.github.com/hse-project or optimize MongoDB for flash and SCM with HSE.

Infographic image comparing Heterogeneous-Memory Storage Engine (HSE) to a legacy storage engine

Features and Benefits

Unleash the performance of your storage-class memory and SSDs

Overcome HDD-based architectures to improve throughput up to six times, latency up to 11 times and SSD endurance by seven times when compared to one leading open source storage engine¹. Most storage engines, and therefore most storage applications, were written for hard drives and never optimized for SSDs, flash-based technologies or storage class memory. HSE offers a way to maximize those legacy applications.

Scale your storage applications to massive scale

Many applications struggle to scale enough for today’s largest workloads. HSE is designed with massive-scale databases in mind, such as billions of key counts, terabytes of data, and thousands of concurrent operations. Integration with MongoDB, the world’s most popular NoSQL database, delivers significant performance improvements, reduces latency and takes full advantage of modern memory and storage technologies. It can also be integrated with other storage applications like NoSQL databases and object stores.

Build on leading-edge storage technologies

Few enterprise applications optimize across the complex stack of different nonvolatile memory hardware. HSE transparently exploits classes of media like QLC 3D NAND flash technology.

Minds of Micron: Steve Moyer on the Open Source Heterogeneous-Memory Storage Engine Project

Learn more about HSE from Steve Moyer, the leader of the team that built it.

Watch video

BLOG

What is a “Heterogeneous Memory” Storage Engine (HSE)?

In case you haven’t already heard, Micron recently released its heterogeneous-memory storage engine (HSE) to the open source community. Our design focuses on providing a solution that makes storage class memory (SCM) and SSDs more performant, with increased effective SSD lifespan through reduced write amplification, all while being deployed at massive scale.

TECHNICAL BRIEF

Micron HSE 2.0: A Step Forward For An Open-Source Storage Engine

Micron’s next-generation, open-source storage engine – HSE 2.0 simplifies deployment and offers more integration options, a more collaborative development environment and enhanced performance

Press Release

Micron Unveils World’s First Open-Source Storage Engine Designed for SSDs and Storage Class Memory

Micron Technology announces the first open-source, heterogeneous-memory storage engine (HSE), designed specifically for solid-state drives (SSDs) and storage-class memory (SCM).

¹Tests performed using HSE v1.7.0 vs RocksDB v6.6.4 using YCSB v0.17.0 and four Micron 9300 SSDs. Data represents the maximum performance from various YCSB workload tests.
² Tests performed using HSE v1.7.0 vs RocksDB v6.6.4 using YCSB v0.17.0 and four Micron 5210 SSDs with and without an X100 SSD. Data represents the performance from YCSB Workload A tests.