 | This articlerelies excessively onreferences toprimary sources. Please improve this article by addingsecondary or tertiary sources. Find sources: "OpenIO" – news ·newspapers ·books ·scholar ·JSTOR(August 2017) (Learn how and when to remove this message) |
 | |
| Industry | Information technology,Data storage,Data processing |
|---|---|
| Founded | 2015 |
| Defunct | 2020 |
| Headquarters | |
OpenIO offeredobject storage for a wide range of high-performance applications.[1] OpenIO was founded in 2015 by Laurent Denel (CEO), Jean-François Smigielski (CTO) and five other co-founders; it leveragedopen source software, developed since 2006,[2] based on agrid technology that enabled dynamic behaviour and supported heterogenous hardware. In October 2017 OpenIO was completed a $5 million funding rounds.[3] In July 2020 OpenIO had been acquired byOVH[4] and withdrawn from the market to become the core technology ofOVHcloud object storage offering.[5]
OpenIO is asoftware-definedobject store that supportsS3 and can be deployedon-premises,cloud-hosted or at theedge, on any hardware mix. It has been designed from the beginning for performance and cost-efficiency at any scale,[6] and it has been optimized forBig Data,HPC andAI.[7]
OpenIO stores objects within a flat structure within a massively distributed directory withindirections, which allows the data query path to be independent of the number of nodes and the performance not to be affected by the growth of capacity. Servers are organized as a grid of nodes massivelydistributed, where each node takes part in directory and storage services, which ensures that there is nosingle point of failure and that new nodes are automatically discovered and immediately available without the need to rebalance data.[8]
The software is built on top of a technology that ensures optimal data placement based on real-time metrics and allows the addition or removal of storage devices with automatic performance andload impact optimization.[9][10] For data protection OpenIO has synchronous and asynchronousreplication with multiple copies, and anerasure coding implementation based onReed-Solomon that can be deployed in one data center or geo-distributed or stretchedclusters.[11][12]
The software has a feature that catches all events that occur in the cluster and can pass them up in thestack or to applications running on OpenIO nodes. This enablesevent-driven computing directly into the storage infrastructure.[13][14]
The open source code is available onGithub and it is licensed underAGPL3 for server code andLGPL3 for client code.
OpenIO claimed in 2019 to have reached 1.372 Tbit/s write speed (171 GB/s) on a cluster of 350 physical machines.[15] The benchmark scenario, conducted under production conditions with standard hardware (commodity servers with 7200 rpmHDDs), consisted in backing up a 38PBHadoopdatalake via the DistCp command.[16] This level of performance marked, according to analysts,[17] the arrival of a new generation ofobject storage technologies oriented toward high performance and hyper-scalability.