Open Standards Make for a Faster 'Shark'
Page 1 of 1
IBM has embraced open standards to boost the data recovery capabilities of its TotalStorage Enterprise Storage Server (ESS), codenamed "Shark."
IBM will add Fibre Channel (FC) support for Shark's Peer-to-Peer Remote Copy (PPRC) disaster recovery and remote backup software, resulting in data transfer up to eight times faster than Big Blue's own proprietary ESCON technology, while reducing channels by a factor of four.
The switch to Fibre Channel "gives us phenomenal performance, up to eight times the throughput of ESCON," says Jim Tuckwell, IBM's ESS marketing manager. And that's just the published data; Tuckwell reports the new FC-based PPRC "has blown through 16 times in the lab...These are real exciting improvements in speed."
The GDPS Storage Manager disaster recovery solution, or "GDPS Lite," functions as an entry-level version of the zSeries GDPS offering that gives the ESS PPRC full disaster recovery automation and data integrity, and is "an excellent way to get started," according to Tuckwell.
PPRC can be run in a synchronized solution up to 103 kilometers, in an asynchronous solution at "virtually unlimited" distances, and in a cascading 2-3 site scenario.
IBM will also release a Copy Services management application interface based on the Storage Networking Industry Association's Storage Management Initiative Specification (SMI-S) for the ESS API. Use of the standard interface is a step toward letting customers and management software vendors develop applications for a single standard instead of developing unique, proprietary interfaces for each vendor, according to Big Blue.
IBM is also unveiling a new Turbo II processor option designed to speed ESS transactions by up to 30% over an ESS base Model 800.
Big Blue also has a new configuration option called Arrays Across Loops that can improve performance for sequential applications up to 1.8 times on a single array.
IBM, CNT Demo Long-Distance Mirroring
IBM will demonstrate the new Shark in an event today with CNT. The two will demonstrate a Fibre Channel over IP data mirroring implementation spanning 1,200 miles, from CNT's new IBM TotalStorage Solution Center (TSSC) in southern New Jersey to CNT's remote demonstration site in Minnesota. The two plan global demonstrations later.
Each location of the demo includes an IBM ESS and a CNT FC/9000 Fibre Channel Director with an integrated UltraNet Edge Storage Router WAN module, which is included in the FC/9000.
The companies claim the technology can potentially slash telecommunications bills for backup services in half.
A large portion of the expense associated with disaster recovery solutions is the cost of telecommunications between the primary and remote data sites, the companies say. The faster Fibre Channel line on IBM's Shark PPRC software results in fewer total connections and a reduced channel infrastructure.
"By eliminating many of the channels previously required to perform automated backup, the Shark improvements can reduce those costs by up to 50%," says Cindy Grossman, director of disk marketing for IBM Storage.
Improved disaster recovery manageability will also allow changes to be tracked to data on the backup volumes while the primary storage is unavailable. These fail-safes help to greatly reduce the time needed to re-synchronize PPRC volumes after switching from the primary to the remote site, according to IBM.
The new ESS enhancements will be generally available Nov. 21. MSRP begins at $250,000-$300,000.
Back to Enterprise Storage Forum