31st International Conference
on Massive Storage Systems
and Technology (MSST 2015)

Sponsored by Santa Clara University,
School of Engineering

Since the conference was founded by the leading national laboratories, MSST has been a venue for massive-scale storage architects, operators, researchers, and vendors to discuss building and securing the world's largest storage systems for high-performance computing, web-scale systems, and enterprises.

May 30th — June 5th

Santa Clara

Santa Clara

MSST 2015 Speaker

Paul Speciale, Scality

Enabling Petabyte-Scale Active Archives through Software Defined Storage

Unstructured data in the form of traditional files and new object payloads is now being generated in massive volumes from both commercial and scientific applications, with over 40 zettabytes (40 trillion GB) of estimated data content produced by the year 2020.(1) High-resolution video data alone can generate over 12 TB per hour for raw (4K) video. Applications in geo-seismic, energy research, medical images and consumer online (cloud based) content sharing—provide a basis for the hundreds of exabytes of new unstructured data being generated on an annual basis. Much of this content will require a form of long-term archival data storage solution as the data is reused, mined and monetized over many years. Regulatory compliance requirements for long-term archiving will also drive a considerable amount of data storage.

The key attributes of an effective petabyte scale active archive solution are centered on massive scalability, high levels of data durability and integrity, multi-protocol online access to data, and transparent migration across platform technology generations without any system or data unavailability. In this session, we investigate how modern scale-out Software Defined Storage (SDS) solutions can enable seamless capacity and performance scaling to support hundreds of petabytes of data with flexible and durable data protection policies, access via a rich set of fast data access methods including legacy file protocols and new object (REST) based protocols, and a pure hardware-agnostic approach to deployment that gives users freedom to choose the most effective platform at each stage of growth. Pure software-defined solutions enable users to scale the system over multiple hardware generations without the need for manual data migrations or downtime.

(1)—Source: IDC Digital Universe (http://www.kdnuggets.com/2012/12/idc-digital-universe-2020.html)

Page Updated April 25, 2017