IBM Software-Defined Storage Technology for Big Data

IBM Corp. unveiled a portfolio of software-defined storage products that deliver improved economics at the same time they enable organizations to access and process any type of data, on any type of storage device, anywhere in the world.

 
One technology in the portfolio, codenamed Elastic Storage, offers performance, infinite scale, and is capable of reducing storage costs up to 90% by automatically moving data onto the most economical storage device.
 
Born in IBM Research Labs, this new, patented technology allows enterprises to exploit – not just manage – the growth of data in a variety of forms generated by countless devices, sensors, business processes, and social networks.

The new storage software is suited for the most data-intensive applications, which require high-speed access to massive volumes of information – from seismic data processing, risk management and financial analysis, weather modeling, and scientific research, to determining the next best action in real-time retail situations.

“Digital information is growing at such a rapid rate and in such dramatic volumes that traditional storage systems used to house and manage it will eventually run out of runway,” said Tom Rosamilia, SVP, IBM systems and technology group. “Our technology offers the advances in speed, scalability and cost savings that clients require to operate in a world where data is the basis of competitive advantage.”
 
Software-defined storage is a set of software capabilities that automatically manage data locally and globally, providing speed in data access, easier administration and the ability to scale technology infrastructures quickly and more cost-effectively as data volumes expand. In addition, these advances can work with any company’s storage systems to provide automated and virtualized storage.
 
Game-Changing Technology
The foundations of Elastic Storage were used for the Jeopardy! television match between IBM’s Watson and two former Jeopardy! champions. For the show, Watson had access to 200 million pages of structured and unstructured data, including the full text of Wikipedia. By using Elastic Storage capabilities, around 5TB of Watson’s ‘knowledge’ (or 200 million pages of data) were loaded in minutes into the computer’s memory.

A key reason these capabilities were chosen for the Watson system that competed on Jeopardy! was its scalability, the architectural limits for which stretch into the thousands of yottabytes. A yottabyte is the equivalent of a data center the size of one million city blocks, which would fill the states of Delaware and Rhode Island combined.

IBM Research has demonstrated that Elastic Storage can scan 10 billion files on a single cluster in 43mn – a technology demonstration that translates into performance for clients analyzing massive data repositories to extract business insights.  
 
At its core, Elastic Storage builds on IBM’s global file system software to provide online storage management, scalable access, and integrated data governance tools capable of managing vast amounts of data and billions of files. For example, it also exploits server-side flash for up to six times increase in performance than with standard SAS disks. This feature recognizes when a server has flash storage and automatically uses that flash as cache memory to improve performance.
 
Elastic Storage virtualizes the storage allowing multiple systems and applications to share common pools of storage. This enables transparent global access to data without the need to modify applications and without the need for additional and often disruptive storage management applications. Since Elastic Storage is not reliant on centralized management to determine file location and placement, customers can have continuous and highly-available access to data in the event of software or hardware failures.
 
For the National Center for Atmospheric Research’s Computational and Information Services Laboratory (CISL), growing data volumes are part of its DNA. The organization, which stores and manages more than 50PB between its Wyoming and Colorado centers, relies on Elastic Storage to give researchers fast access to vast amounts of diverse data.
 
“We provide computational, educational, and research data services for geosciences to more than 1,000 users at more than 200 different sites,” said Pamela Gillman, manager, data analysis services group, CISL. “The IBM global file system software has enabled scalable, reliable and fast access to this information. That has dramatically improved the performance of the different functions, as well as the organization as a whole.“
 
A key component of Elastic Storage is its ability to automatically and intelligently move data to the most strategic and economic storage system available. Through policy-driven features and real-time analytics, for example, it can automatically move infrequently-used data to less expensive low-cost tape drives, while storing more frequently-accessed data on high-speed Flash systems for quicker access. Such policy-driven features can provide cost savings of up to 90%.
 
In addition, the software features native encryption and secure erase, which ensures that data is irretrievable to comply with regulations such as HIPAA and Sarbanes-Oxley.
 
Through its support of OpenStack cloud management software, Elastic Storage also enables customers to store, manage and access data across private, public and hybrid clouds for global data sharing and collaboration. In addition to supporting OpenStack Cinder and Swift access, it supports other open APIs such as POSIX and Hadoop.
 
While traditional storage systems must move data to separate designated systems for transaction processing and analytics, Elastic Storage can automatically balance resources to support both types of application workloads, including Hadoop based analytics. This speeds analysis and eliminates the costly and time-consuming process of producing duplicate copies of data.
 
Elastic Storage software will also be available as an IBM SoftLayer cloud service later this year.
 
About IBM Big Data and Analytics
Each day, we create 2.5 quintillion bytes of data generated by a variety of sources – from climate information, to posts on social media sites, and from purchase transaction records to healthcare medical images. IBM believes that data is emerging as the world’s newest resource for competitive advantage, and analytics is the key to make sense of it. The company is helping clients harness big data and analytics to provide insights needed to make better decisions, create value, and deliver that value to clients and society.

IBM has a broad portfolio of big data and analytics technologies and solutions, spanning services, software, research and hardware.