IBM introduces tomorrow's tape storage offerings today that are capable of handling a massive 2.7 exabytes using new "shuttle" technology.
IBM has announced new tape storage and enhanced archiving, deduplication offerings designed to help clients efficiently store and extract intelligence from massive amounts of data.
The amount of information generated is increasing dramatically each year driven by armies of sensors, mobile devices, social networks, cloud computing, and public sources of information like the Web. At the same time, demand for storage capacity worldwide will continue to grow at a compound annual growth rate of 49.8 percent from 2009-2014, according to IDC (1). Clients are requiring new technologies and ways to capitalize on the growing volume, variety and velocity of information known as "Big Data," according to IBM.
The company continues to invest in the development of tape storage as an inexpensive, efficient way to store massive volumes of data. For example, IBM is today detailing seven significant improvements to its portfolio of tape and other storage archiving products, including the industry's first tape library system to provide over 2.7 exabytes of automated, low cost storage--enough to store nearly three times all the mobile data generated in the U.S. in 2010 (an exabyte is 1 followed by 18 zeroes) (2).
IBM System Storage TS3500 Tape Library is enabled by a new, IBM-developed shuttle technology--a mechanical attachment that connects up to 15 tape libraries to create a single, high capacity library complex at a lower cost. The TS3500 offers 80 percent more capacity than a comparable Oracle tape library and is the highest capacity library in the industry, making it ideal for the world's largest data archives (3).
For example, the National Institute of Water & Atmospheric Research (NIWA) in New Zealand uses two TS3500 Tape Libraries to support a sophisticated POWER supercomputer used to tackle challenges in such fields as energy, aerospace, weather and climate modeling. The libraries together hold five petabytes of data, the equivalent of more than 1,000,000 DVDs. This means that if a DVD were written to the tape libraries at the rate of one every minute, it would take more than two years before their capacity was exhausted.
IBM in 1952 pioneered the use of magnetic tape for data storage, which is one of the 100 Icons of Progress in IBM's century-long history. IBM is announcing the IBM System Storage TS1140 Tape Drive, which holds two million times more data than IBM's first tape drive (4). The TS1140 employs fewer and more efficient components developed with IBM Research, enabling it to use up to 64 percent less energy and can deliver up to 80 percent more performance than a comparable Oracle drive, so clients can improve productivity at a lower cost (5).
Tape storage is gaining new opportunities as a low cost, energy efficient way for companies in industries like media and entertainment and healthcare to create huge archives of digital assets and information being driven by Big Data. In fact, the amount of digital archives stored to tape is set to experience a six fold increase from 2010 through 2015, according to Enterprise Strategy Group (6). IBM is well positioned to capitalize on this opportunity as the tape storage market share leader with over 40 percent revenue share--more than twice as much share as the next competitors HP and Oracle, according to IDC (7).
IBM has announced file system access to select IBM tape libraries with the IBM Linear Tape File System Library Edition (LTFS LE), invented by IBM Research to provide a simple, cost efficient way to access and manage massive archives of data and digital assets. LTFS clients can now more efficiently index, search, retrieve and share data stored on Generation 5 LTO tape, an open tape storage format.
IBM has a clear edge over storage vendors like EMC that don't support or develop tape. Tape and disk can be used together to deliver clients tiered storage that enables them to store data within the different tiers based on their data priorities. For example, IBM's Scale-out Network Attached Storage (SONAS) system and Information Archive both use policies to automatically migrate data to the tape storage tier. This can reduce total cost of ownership for clients by up to 40 percent in the long-term, according to an internal IBM study (8).
Klinikverbund Suedwest GmbH is an association of clinics and hospitals in Southwest Germany with 4,200 employees caring for more than 250,000 patients each year. Working with IBM Business Partner Profi AG, the Association installed the IBM XIV storage system with IBM Tivoli Storage Manager to work with its existing IBM System Storage TS3310 Tape Library. "By using both IBM tape and disk technology, the Association has lowered storage costs significantly," said Dr. med. Gunther K. Weiss, M.Sc., managing director, Klinikverbund Suedwest GmbH. "With these different storage tiers at our disposal, we can ensure that our clients and employees have secure, ready access to their data."
IBM also announced enhancements to SONAS, which scales to over 14 petabytes of clustered storage and is based on technology invented by IBM Research. SONAS now offers double the throughput -- or the amount of data processed by the system during a given period of time -- of its predecessor and faster response times (9). The technology, which will provide support for commonly used anti-virus applications, is designed to offer protection from viruses and malware threats with the ability to scan archived data and isolate or delete compromised files. In addition, SONAS will now support an open standard protocol called NDMP, enabling clients to backup and protect large amounts of data in SONAS using ISV applications that support NDMP.
IBM is also improving the IBM Information Archive for Email, Files and eDiscovery, a pre-installed, pre-configured archiving solution. With integrated hardware, software and services, the technology can reduce the cost of installing and implementing a storage archive by up to 70 percent compared to starting with piece parts and can be deployed in a few days, according to IBM's internal measurements (10).
Finally, IBM is announcing new or enhanced tape virtualization offerings for mainframe or open storage environments to provide clients improved access to and protection for their data in the virtualized datacenter. These include:
- The addition of a many-to-many replication feature to IBM System Storage TS7650 ProtecTIER Deduplication Solutions that will allow enterprises with multiple data centers to automatically replicate backup data between locations so multiple copies of critical data can be stored and quickly restored if needed. By stripping out duplicate data before it's replicated, ProtecTIER customers have experienced a reduction in bandwidth needed to send the data of up to 95 percent or more, reducing the amount of storage needed to archive and protect data (11); and
- Enhancements to the TS7700 Virtual Tape Library family include doubling the number of virtual tape cartridges it contains to 2 million and leveraging IBM POWER7 technology to significantly improve performance, delivering 70 percent better performance than a comparable Oracle offering (12).
"As part of its $6 billion annual investment in R&D, IBM continues to develop innovative storage technologies for smarter computing," said Brian Truskowski, general manager, IBM System Storage and Networking. "The data protection and retention solutions IBM is announcing today provide clients with a comprehensive approach to protecting and accessing their most valuable asset--their data."
For more information on IBM storage, visit http://www-03.ibm.com/systems/storage/. For more information on IBM Smarter Computing, visit:
http://www-03.ibm.com/systems/data/flash/smartercomputing/. Also visit www.storagecommunity.org, a new online storage community launched today and sponsored by IBM. Participants can network with peers, share best practices and dialogue with storage experts.
1. IDC, Worldwide Enterprise Storage Systems 2010-2014 Forecast Update: Better Expectations on 2010 Growth, No Changes to Long-Term Outlook, Doc # 224618, August 2010.
2. Data traffic on cell phones in the U.S. reached an exabyte in 2010, according to Chetan Sharma Consulting. Comparison assumes 3:1 compressed data.
3. As compared to the Oracle SL8500 max tape library complex with native T10000c, which offers a max capacity of 500PB. The TS3500 max tape library complex with native TS1140 offers a max capacity of 900PB, when maximizing cartridge capacity. SL8500 source: http://www.oracle.com/us/products/servers-storage/storage/tape-storage/034341.pdf.
4. The TS1140 holds 4,000,000MB. The IBM 726 Magnetic Tape Unit held 2MB.
5. As compared to the Oracle T10k-C. The TS1140 operates at 51W and at 24W standby power. The Oracle T10k-C operates at 67W max power. The TS1140 max throughput is 650MBps. The Oracle T10k-C max throughput is 360MBps. Assumes compressible data.
6. Briefs: NERSC: Proving Tape as Cost-Effective and Reliable Primary Data Storage, Enterprise Strategy Group, December 2010.
7. Total branded tape storage revenue share, IDC Branded Tape Pivot Q4CY10.
8. Based on an internal IBM study.
9. SONAS single interface node doubled the available bandwidth with dual active 10 GbE interface ports, while code optimization improved pathlength efficiency.
10. Based on IBM's internal measurements.
11. ProtecTIER's deduplication reduces storage requirements 20:1 per IBM production client data.
12. The TS7700 delivers up to 900MBps throughput compared to Oracle StorageTek Virtual Storage Manager's 640MBps throughput.