During the last IT Press Tour (the 14th), in the Silicon Valley (December 1-5, 2014) we met several companies in different categories (Cloud, Storage and Big Data).
The last company that we met during the fourth day was Scality, that produces a storage software product called the RING for large scale data storage.
Scality was founded in 2009 by a team of entrepreneurs and technologists (Jérôme Lecat, Giorgio Regni, Daniel Binsfeld, Serge Dugas, and Brad King) and actually it has more than 90 employees and more than 50 production customers.
When the Scality team talked to the initial base of potential customers, the customers wanted a system that could “route” data to and from individual users in the most scalable, efficient way possible.
Leo Leung (VP Corp. Marketing) has explain the company origin their product: Scality RING.
The Scality RING is a software-based storage solution that runs on standard x86 servers and standard Linux with no kernel modifications. Unlike traditional storage, the RING is a software-based storage that is completely decoupled from hardware. The hardware agnosti feature enables flexibility in your choice of x86 servers, your mix of flash, SSD, HDD and newer media, to match the appropriate hardware form factor for your environment. And also give the flexibility in using industry standard servers (or commodity servers): the RING has been deployed across the world with most major server vendors including HP, Dell, IBM, Cisco, SGI and Supermicro.
The RING uses a object storage core for scalable data management, a peer-to-peer architecture for reliable routing (based on the MIT CHORD algorithm), and shared-nothing parallelism for performance. It is designed for hundreds of petabytes of data, hundreds of billions of files, and continuous availability at scale, with the ability to serve the majority of storage workloads via file, object, and VM-based interfaces.
The RING is an always-on infrastructure: instead of small-scale design points like active/passive failover, RAID, backups and offsite backups, the RING is designed to remain available even with multiple failures in the system at any given time, prodiving data protection including Geo-Redundancy functions.
Scality RING can solve significant challenges:
- Capacity growth: petabyte-scale with no new admins and scale-out architecture
- Legacy Storage silos administration and costs: runs 80% of storage workloads and it’s easier and cheaper in maintenance
- Demand for “Always on”: 100% uptime and durability and self-healing
- Cloud economics: software-based and hardware-agnostic
Scality’s latest release, RING 5.0, allows enterprises to run multiple type of storage workloads: files, objects and even Virtual Machine (VM) storage. Either on-premises or as a service and with linear performance and scalability across workloads. Scality delivers a unified storage approach that enables data consolidation and can increase utilization and economies of scale, and eliminate storage silos.
With the latest release there is also an OpenStack integration: Scality is already compatible with Cinder and Swift, but with non-intrusive drivers is possible facilitate OpenStack customers support and provide future evolution, like on-going integration with Dashboard, Telemetry, Identity, Networking…
And there is also new Open Source contribution: an Open Source Scality REST block driver that presents standard local block devices on Linux servers, installed as a Linux kernel driver.
For more information about Scality see also:
- Software Defined Storage–Scality Unifies File Object And VM Storage
- Scality likes object storage, puts a RING on it… for the fifth time
- HP slips on Scality’s RING, plans to flog it with ProLiant servers
Disclaimer: I’ve been invited to this event by Condor Consulting Group and they have paid for accommodation and travels, but I’m not compensated for my time and I’m not obliged to blog. Furthermore, the content is not reviewed, approved or published by any other person than me.