By Dave Simpson
Analysts view EMC's acquisition of Legato, which is expected to be finalized in October, as a win-win for both companies, and most analysts see it as a positive deal for end users—with a few caveats.
"We expect end users to win with this deal as long as EMC doesn't apply its legacy rules to pricing and lets Legato's development and sales teams show them how to sell software on its own value basis rather than as an adjunct to hardware," says Arun Taneja, consulting analyst and founder of The Taneja Group. "End users will see a much fuller software portfolio...to manage data from inception to deletion."
With the acquisition, EMC gains a variety of software product lines, including backup and restore, server-based replication, application monitoring and clustering, media management, hierarchical storage management (HSM), archiving, and e-mail and content management software.
However, EMC gains a lot more than products. "This helps EMC to establish a substantially broader base of customers, and in that sense it's similar to the EMC-BMC deal," says Richard R. Lee, president and CEO of The Storage Consulting Group. "It's not so much a case of acquiring technology as it is acquiring customers, although Legato's software will fill some holes in EMC's portfolio." Lee adds that the services portion of Legato's business (which accounts for about 54% of Legato's total revenues) will also be key for EMC.
Legato's software is expected to fill in the holes in EMC's portfolio.
But as analysts and EMC's competitors were quick to point out, EMC will have to deal with software integration issues and product overlap. Although EMC officials maintain that there is little product overlap (see table), Lee notes that "EMC will have to figure out what products to keep and which to get rid of."
EMC has not provided details on its plans for integration of the two companies' product lines, other than stating that "Legato has been a long-time API partner and some level of integration already exists." Details are expected at the time the acquisition closes.
EMC officials are clear on one area of product overlap—backup and restore. President and CEO Joe Tucci has stated that, "over time we will take EDM [EMC Data Manager] technology and embed it into [Legato's] NetWorker. Our customers can count on EMC to continue to support and moderately enhance EDM." Eventually, EMC will give EDM users a free upgrade to NetWorker, which will become the company's primary backup-and-recovery offering. EDM is expected to remain a Symmetrix-centric backup/restore product.
Anders Lofgren, senior industry analyst at Forrester Research, says that the software that Legato got in its acquisition of OTG Software is even more important to EMC's strategy—and, possibly, to end users—than the well-known NetWorker product line. That software includes HSM and archiving products, as well as content/records/e-mail management software.
"The OTG software is key to EMC's 'information life-cycle management' strategy," says Lofgren, "and it will help them to sell more Centera boxes." For end users, the key advantage is one-stop shopping. "Previously, users had to buy point products, often from different vendors and for single applications. EMC can now provide a storage management solution that runs across multiple applications, with the same underlying storage management structure."
Aside from sorting out the product lines, perhaps EMC's biggest challenge is to convince end users and its partners that the company is an open software vendor that is not bent on driving sales of its hardware. EMC has repeatedly stated its goal of deriving 30% of total revenues from software. That figure is currently at about 23%, and the Legato acquisition will only bring it to approximately 25%, according to EMC officials. Of EMC's total software revenues, about 33% comes from open (e.g., not Symmetrix-specific) software.
"If EMC truly welcomes Legato's development and sales/marketing teams and doesn't squash them with hardware-centric attitudes, this could be a major step toward EMC's strategic desire to become a heterogeneous storage management software player," says Taneja.
Analysts say that the acquisition of Legato will put EMC on a head-on collision course with Veritas, which has also made recent acquisitions to broaden its portfolio beyond just storage software (see "Veritas continues push beyond storage," p. 8), as well as with archrivals such as Computer Associates and IBM/Tivoli.
The Storage Consulting Group's Lee sees a four-way horse race developing for leadership in the storage management software market, with the primary competitors being EMC, Veritas, IBM/Tivoli, and CA. Other analysts include Hewlett-Packard in the list.
"All of the vendors are going for portfolios that range all the way from backup and restore to application performance management," says Lee, noting also that "there's a real movement among end users to go back to one-stop shopping for management software."
In addition to Legato's products and customer base (Legato claims more than 31,000 customers), the acquisition will give EMC access to Legato's approximately 400 channel partners, which Taneja says is critical for EMC to compete with Veritas. "EMC gains access to a large number of channel partners that it would have had difficulty wooing, given its past reputation," he says. However, the company is expected to lose the revenue from large Legato software resellers such as Hewlett-Packard and Sun.
The EMC-Legato acquisition (a stock-only deal) was valued at $1.3 billion at the time it was announced. EMC plans to run Legato as a software division (similar to IBM's Tivoli division) and is expected to keep the Legato brand name.
In a separate announcement earlier this summer, EMC entered into a partnership with former competitor BMC Software. BMC will resell the portions of EMC ControlCenter (ECC) that are not hardware-specific, which includes seven modules: StorageScope, SAN Manager, Automated Resource Manager, Common Array Manager, SAN Architect, Workload Analyzer, and AutoAdvice. According to the Gartner consulting firm, all of these packages fall into the general product category of storage resource management (SRM).
BMC will sell ECC through its MarketZone Direct channel/reseller program.
EMC, in turn, acquired BMC's Patrol Storage Manager (PSM) software assets. (BMC ceased development of PSM in February.) However, neither EMC nor BMC will continue to develop PSM (although EMC will support the installed base through January 2005), and PSM will not be integrated with ECC. (EMC will, however, integrate ECC with non-PSM elements of BMC's Patrol family.)
"This isn't about filling EMC's portfolio holes," says Barry Ader, director of software product marketing at EMC. "It's about migrating PSM customers to ControlCenter."
The deal marks EMC's first reseller partnership with an independent software provider. Financial details of the deal were not disclosed.
EMC enhances DMX arrays
Next month, EMC will introduce a number of new features and arrays in its DMX line that may turn some heads. For example, the company will debut non-disruptive capability across the product line. In addition, EMC will for the first time support iSCSI and will offer an asynchronous version of its SRDF software.
The non-disruptive capabilities come via a new version of EMC's Engenuity operating software and allow users to perform tasks such as hardware/software upgrades, reconfigurations, and service-related operations without bringing down the array. Previously, users had to buy extra host bus adapters (HBAs) and expensive path-fail-over software to handle those operations non-disruptively, according to Chuck Hollis, vice president of storage platform marketing at EMC. The new version of Engenuity, which EMC claims also provides a 20% performance improvement, runs on all DMX arrays and is available as a free upgrade.
Also next month, EMC will ship a multi-protocol card for its DMX arrays that supports 2Gbps FICON, Gigabit Ethernet and, for the first time in a high-end array, iSCSI. The company will also provide an iSCSI "readiness-assessment" service. The multi-protocol card has up to four ports and can be configured with any mix of interfaces.
Hollis says that the iSCSI support is only for server-to-storage connections (as opposed to WAN connections), providing a less expensive alternative to Fibre Channel. EMC has tested the iSCSI connection with HBAs from a variety of vendors.
The Gigabit Ethernet connection is for applications such as remote replication, and eliminates the need for expensive Fibre Channel-to-IP conversion boxes that can cost anywhere from $50,000 to $200,000, according to Hollis.
Also on the replication front, EMC will ship an asynchronous version of its synchronous SRDF software, called SRDF/A. The main advantage of asynchronous replication versus synchronous replication is the ability to replicate over longer (global) distances, but EMC also claims that SRDF/A will provide up to a 30% savings in telco costs compared to other asynchronous replication products due to bandwidth-saving techniques.
EMC will also introduce EMC Snap local replication software, which is pointer-based (as opposed to full-copy replication software such as the company's TimeFinder). Although common on midrange arrays, Snap represents EMC's first pointer-based replication code for its high-end arrays. EMC claims a 70% capacity savings vs. TimeFinder.
Finally, EMC will introduce two new DMX arrays: a low-end, 4-controller version of the rack-mounted DMX800 that costs 30% less than the previous low-end DMX800 ($284,000 for 584GB of raw capacity), and a new high-end model, the DMX3000. The triple-bay DMX3000 provides twice the capacity and performance of the DMX2000.