HP Acquires Cluster File Storage Software Vendor IBRIX

Today, HP announced that they will be acquiring the startup cluster file storage software firm IBRIX in a deal with few disclosed details. Congrats to HP for making another interesting acquisition in the clustered storage space, following their Feb. 2007 acquisition of Polyserve and their October 2008 acquisition of Lefthand. IBRIX makes software that allows users to cluster many nodes of storage (either industry standard servers or iSCSI/FC SAN arrays) for use in high performance global namespace file storage.  IBRIX’s customer base includes some 175 firms, generally in the high performance compute (HPC), video rendering/animation, and oil and gas exploration spaces.

This is an interesting move for HP who points to building share in HPC, content depot/cloud storage and archiving as key goals for this acquisition. Clearly those are high growth markets, so it’s no surprise that HP is interested in accelerating their traction there. But, it is surprising to see archiving and content depot listed as top level goals given that IBRIX doesn’t have WORM capability or object orientation. Additionally, you have to wonder what this says about Polyserve given that they were a maker of high scale global namespace file storage, and is now the basis for HP’s current cloud offering, XDS. There seems to be some overlap there, and it will be interesting to see what lines they draw to delineate use cases for customers.

Another area of concern here is the significant partnerships that IBRIX has with HP rivals EMC and Dell.  IBRIX was the original software used for EMC’s HULK scale out hardware offering before they completed the MAUI software that combined with HULK to make Atmos, EMC’s cloud storage offering. There are still a large number of users with highly complex environments that have IBRIX software mated to EMC as well as Dell hardware as back end storage; HP will have to be careful to avoid disrupting those power user customers as they develop their strategy for moving forward with IBRIX.

As with Polyserve and Lefthand, IBRIX is software only, so it makes sense for it to be owned by a major hardware player that can sell it as a comprehensive solution including hardware and services, rather than having customers cobble solutions together themselves. But, it will be tricky for HP to determine how to support existing solutions with gear from competitive vendors, how much to combine it with other software versus leaving it as a standalone product, and how it can fit with HP’s broader vision of storage.

On the positive side, HP has many back ends that could fit different types of IBRIX deployments, from Proliant servers to Lefthand iSCSI storage to EVA mid range SAN arrays, so the likelihood is that they have the right raw materials to make it work. On the negative side though, HP now has a very broad lineup of storage products, some internally developed, some acquired and some delivered through partnerships, and maintaining a consistent vision and value proposition that meet specific customer requirements may be a challenge.

By Andrew Reichman

Check out Andrew's research

Green Storage Has Limited ROI, But Supports Overall Efficiency

I care deeply about the environment, certainly more than I care personally about money, so it pains me to say that in most cases, making storage decisions based on power expenditure alone is not rational behavior. The world is driven by economics, and the stark reality is that the cost of power is only a drop in the bucket compared to the amount organizations spend on acquiring and managing their enterprise storage systems. Maybe someday a consumption tax or cap and trade system will tip the balance towards more responsible consumption of non-renewable resources, but in the meantime, the pricing of power (especially in the US) doesn’t give much economic incentive for good behavior. In fact, according to a report Forrester published recently, the amount of money typically spent on electricity to power and cool a TB of storage is only about 1% of the cost of buying that TB of storage (or about 4% of the annualized cost of buying that storage given that you only have to buy the TB once every 3-5 years but you power it every year). So, unfortunately for the environment, power cost itself doesn’t provide a very strong incentive for storage efficiency.

Fortunately though, the things that enterprises can do to reduce their power consumption costs are often the exact same things they can do to reduce the capital and operating expenses of their overall storage environment. Focusing on improving utilization (measured as the quantity of data written divided by the quantity of storage on hand) and increased usage of dense drives are the most straightforward and effective ways to reduce hardware acquisition costs as well as power consumption. There are many ways to achieve these objectives such as thin provisioning, reporting and reclamation to improve utilization, and tiering or wide striping to enable more use of dense drives. Whatever the motivation — economic, altruistic, or a combination of both — organizations that put significant focus on their utilization and dense drive ratios are likely to spend less money and be greener at the same time. And that’s good for everybody.

By Andrew Reichman

Check out Andrew's research

HP Acquires LeftHand Networks, Virtualized iSCSI Storage Vendor

Today, HP announced the intention to acquire iSCSI storage vendor LeftHand Networks for $360 Million in cash. LeftHand makes virtualized storage array software that can turn just about any server, PC or subset of disks on a device into a full featured IP based SAN array with features like space efficient snapshots, distance replication support and thin provisioning. LeftHand has a tight partnership with VMware, and does both standard virtual server attach to SAN, as well as some innovative virtual appliance models that turn the physical disks on board a VMware server into an array. The LeftHand software can span across multiple physical devices, allowing for combinations of virtual appliances and dedicated servers.

It’s no surprise to see LeftHand get snapped up by a major storage vendor as they are one of the few remaining independent storage vendors in a market where it’s hard to be small and go it alone. It’s a particularly good fit for HP though for a number of reasons. For one, about half of LeftHand's 3000+ installs use HP servers as a back end. HP states that they are committed to supporting existing customers regardless of what platform they are on, but they are eager to take advantage of their strength in delivering industry standard servers as the go forward back end for LeftHand. This is also a good thing for customers who will benefit from the simplicity of a well integrated packaged solution including hardware and software. The acquisition also gives an immediate boost to HP's story around advantages for VMware attach, which is currently one of the hottest battlegrounds for all storage vendors.

What's more, like PolyServ, the clustered NAS firm bough by HP, LeftHand is at its core a software company. This means that HP can continue selling the current LeftHand product family, and eventually take advantage of the capabilities throughout the rest of the HP StorageWorks family. Existing HP storage products can use the help as the iSCSI offerings are weak, they have no true thin provisioning, snapshot count and manageability can use a boost and the overall EVA architecture would benefit from more virtualization and flexibility. The LeftHand technology stack should help to accelerate HPs roadmap in these key areas, helping them improve current offerings and at the same time build towards the virtualized, scale-out storage that is likely to be the architecture of choice in most datacenters in the near future.

By Andrew Reichman

Check out Andrew's research

Oracle Fires A Booming Shot Across The Bow Of The Storage Vendor Establishment

Yesterday at Oracle OpenWorld, Larry Ellison announced the database giant's first foray into the hardware realm, unveiling the HP/Oracle Database Machine, branded as Exadata. The Exadata system is a combination of Oracle's 11g database engine using Automated Storage Management (ASM) to manage a grid of HP Proliant servers with 12 SATA or SAS drives each, connected to database servers via InfiniBand. Oracle separated the database processing to allow parts of queries to be run on the storage servers, delivering query results to the database servers rather than whole tables, reducing the amount of I/O passing between servers and storage. There is no interconnect among the storage servers, instead ASM manages read and write requests in parallel among all the storage boxes. The storage server grid allows replacement of failed nodes and ongoing scalability to be automated, and growth to be theoretically unlimited. Oracle claims significant performance increases over traditional architectures and dramatic price improvements at the same time, given the industry standard x86 server architecture and the wide use of dense SATA drives.

Leave it to Larry Ellison not to play politics and come right out with an announcement that goes right at storage vendor giants, many of which are platinum sponsors at his event. While the announcement was primarily aimed at data warehouse applications, with Teradata and Netezza listed as primary competitors, there were specific references to data warehouse implementations on traditional storage products, as well as strong capabilities with OLTP database workloads. EMC, Hitachi, and NetApp systems were described in the announcement with performance capabilities 10-30x lower than the new Oracle appliance, with much higher price tags. Larry's own company, Pillar Data made announcements at OpenWorld touting new strength in Oracle specific tuning and configuration aimed at optimizing performance, but he did not shy away from vigorously promoting a new approach to storage for Oracle that leaves dedicated storage vendors struggling for relevance.

This is the most dramatic announcement in storage in a decade, and could have the effect of significantly lowering TCO and improving performance across database applications of all kinds. Oracle customers, especially those with large data warehouse needs should evaluate their architecture and determine whether spending money on intelligent storage arrays still makes sense given the performance and automation that can be gained from using the Oracle HP platform. Time will tell if Oracle can execute on this ambitious offering, but if they can deliver on their promises, all the major storage vendors should be watching their backs.

By Andrew Reichman

Check out Andrew's research

Xiotech Zigs When Other Storage Vendors Zag

Bucking the ‘hardware is a commodity’ trend seen widely in the storage industry these days, emerging vendor Xiotech recently announced their Emprise line of storage arrays based on a modular component called Intelligent Storage Element (ISE), housing drives in sealed DataPacs designed to dampen vibration, prolong lifespan, and reduce the incidence of “no fault errors.”

The ISE relies on a custom enclosure architecture that mounts drives in an opposing fashion to cancel vibration and uses advanced materials and design to improve cooling and IOPS performance. While the drives inside are standard Seagate HDDs, they use specialized firmware to allow for advanced error management and give ISEs the ability to manage around errors, bypassing bad sectors without having to swap out individual failed drives. This has the effect of increasing the size of the base unit of storage, from the individual drive to the ISE, allowing storage admins to spend their time on higher level data management while base level functions like striping, drive, and spindle management are automated.

Technology features aside, the Xiotech solution drives business impact for users. Because of the improved reliability of the new architecture, the vendor offers a five year warranty standard with Emprise arrays, contrasted with typical warranties of three years in the storage industry. This relieves customers from the burden of costly and complex migrations after only three years of service, or the prospect of paying around 15% of total array purchase price for annual maintenance — either of which can add up very quickly in terms of overall IT expenditures. Combine this cost avoidance with the claims for easier management for an ISE based system and the result is a significant reduction of overall operating expense related to storage. Xiotech, a relatively small fish in the ocean of sharks that is the storage industry, gains a serious product differentiator with this announcement — one that can directly translate into reduced deployment and management costs for storage; a timely attribute in this time of shrinking budgets and ever growing data.

By Andrew Reichman

Check out Andrew’s research

EMC Bolsters Data Protection Reporting Capabilities With Acquisition Of Partner WysDM

News emerged Friday regarding EMC’s acquisition of data protection reporting software maker WysDM . The move is not entirely surprising due to several years of partnering arrangements between the two companies. WysDM, the New York City based independent software vendor, brings significant enterprise credibility gleaned from a Wall Street pedigree, with several founders coming not from Silicon Valley, but from the technology organizations of major investment banks. WysDM offers a strong portfolio of backup and primary data reporting tools — an area that is sorely lacking in most data centers today. With all storage environments growing at a significant pace, companies need to know how much data resides in what store, what the backup and replication scheme is for each piece of data, and must have the ability to measure the performance of all of the elements of a complex data storage environment. In most environments, backup job success rates are abysmal, policies are nearly impossible to audit, the ability to meet backup windows are under severe pressure, and decision-makers generally don’t have the tools to identify what the bottleneck truly is.

WysDM offers tools to assist with all of these problems, and EMC is likely to put them in the hands of more customers. Symantec has offered the Backup Reporter module for their NetBackup product for more than a year, and CommVault has come to challenge the big players in this market due in no small part to advanced data reporting capabilities. EMC’s Legato and IBM Tivoli have lagged in this area, and it therefore makes sense for EMC to take on the WysDM product, tried and tested as it is in EMC environments. This move will likely prompt IBM to take a good hard look at acquiring WysDM competitor Bocada who they have partnered with. EMC needs to proceed carefully with integration to avoid diluting the strong WysDM interoperability with data protection suites offered by Symantec and IBM.

What is the upside for IT professionals? This is likely to be good for both EMC and consumers as data protection reporting is a key function, directors of enterprise storage environments would be remiss not to have this high on their priority list. If EMC manages this acquisition well, the vendor will use the WysDM technology to give a shot in the arm to the reporting capabilities of their overall storage resource management portfolio, which continues to lag behind that of competitors.

By Andrew Reichman

Check out Andrew’s research

Startup Ocarina Networks Proposes Ground-Breaking Data Footprint Reductions

I met with Ocarina Networks founder Carter George for the first time this week and was genuinely excited about the dramatic data reduction that their cutting edge technology promises. Ocarina aims to reduce the number of bytes that go into file storage through a combination of data deconstruction, deduplication, and compression. Deconstruction breaks complex file types into basic components, deduplication spots block level similarities and replaces them with hashes, and compression squeezes the de-duplicated components beyond what traditional compression techniques can with more complex whole files.

The heavy lifting is done by an out-of-band appliance that applies post processing techniques to create size reduced shadow copies of data with “bit for bit consistency checks” to prevent corruption or loss of data. The key to the approach is a “file type aware approach,” using specialized algorithms and techniques for files like images or text. George claims that the process can reduce the size of most data types by 10X and can even yield a 3X reduction on JPEGs that typically see little benefit from deduplication techniques. For web 2.0 firms that are looking at growing their storage by multiple TB a day, a capability like this would be game changing by reducing their growth costs by leaps and bounds.

There is considerable market resistance to deduplication of primary data, where the industry has long relied on creation of multiple copies to guarantee the protection and availability of critical data. If you are an IT operations professional in the banking or financial services sector, don’t expect your peers to adopt Ocarina’s cutting edge approach any time soon. But, if your firm has less stringent SLAs around data loss and availability and faces staggering growth from day-to-day operations, this could be the missing link that allows you to stay ahead of competitors and offers more rich services to customers without breaking the bank on storage capacity. I can’t speak to the viability of technology approach yet, but suffice it to say that if they can deliver what they promise, this could dramatically shake up the enterprise data storage world and move the industry along the path to keeping up with the astonishing but widely accepted forecasts for growth of file data.

By Andrew Reichman

Check out Andrew’s research

IBM Buys XIV, Maker Of Grid-Like Storage Systems

On December 31, 2007, IBM announced closing the acquisition of Israel based storage systems vendor XIV for an undisclosed amount, but speculated to be in the $300-350MM range. This deal is interesting in terms of technology, as well as politics. Regarding politics, XIV’s Chairman Moshe Yanai was a pioneer at EMC, personally designing the Symmetrix system that continues to be a high-end storage market leader as today’s DMX systems. IBM’s investment in his technology and his company is a validation of the individual who EMC cast aside, as well as the product approach that XIV is taking in the storage market.

The technology itself is an example of a grid-like or clustered storage system that uses industry standard x86 servers with dense drives that are clustered together to form a cohesive storage system. With such an architecture, costs are kept low as there are no custom chipsets or wiring, scalability is easier as the system automatically rebalances across the new nodes, and scaling up improves aggregate performance as the addition of nodes adds processing power, cache and I/O throughput. For many workloads, this type of storage system appears to be the future of storage, offering lower acquisition cost, increased flexibility of data management, massive scalability and much easier management. IBM is not the first vendor to move towards such an architecture, with companies like Isilon, Compellent, Lefthand Networks, 3PAR, EqualLogic (recently acquired by Dell) and Network Appliance offering elements of this type of architecture today. Google, Amazon, Cleversafe and Iron Mountain have similar offerings that they intend to use for software-as-a-service offerings. Even the leader in traditional storage systems appears to be moving in this direction with their Maui/Hulk announcements and their purchase of backup service provider Berkley Data Systems.

IBM has lots of moving parts in their current storage portfolio, with LSI and Network Appliance acting as key partners in bringing current products to market. While there is definitely tremendous potential from moving to a clustered architecture, execution will be critical. Most important will be enabling file access to XIV technology, which does not support CIFS, NFS or clustered/distributed file system capabilities natively, access methods that are critical to Web 2.0 and rich media storage applications. Also, creating a clear strategy of where XIV based technology will make sense for customers compared to existing offerings, and building a robust ecosystem of application partners to integrate with this technology will be critical for success. So far, IBM has shown the strength of the XIV technology, what they need to do next is present a clear vision of how it will fit with their future offerings.

By Andrew Reichman

Check out Andrew's research

3PAR, HP, And VMware Team Up To Promote A Cross Functional Best Practice Architecture, 3CV

On December 10th, 2007, 3PAR announced a partnership with HP and VMware to promote a high performance infrastructure architecture that a group of large customers developed, implemented, and validated. Customer references include marquee companies Deutsche Bank, Hilton, Savvis, and Ariba. Coined “3CV,” which stands for “3” Par, HP “C” Class Blade Center, and “V” Mware, the architecture is said to reduce total hardware acquisition cost, cut power consumption, and increase the agility of provisioning, as well as enable consolidation through security at the virtualization layer rather than through physical separation.

Since a solution like this is inherently cross functional, multiple vendors must cooperate to put a complete and effective solution in place. In general, this is difficult for vendors that don’t want to play favorites among the arch rivals that share space in their partner ecosystem. Without clear best practices that choose one product over another, each customer is left to trial and error to determine which of the multiple diverse layers of physical and virtual infrastructure, software, and middleware work well together and will best solve their problems.

This announcement seems to be a step in the right direction, with leading edge customers that take the risks to be first to market opting to share their experiences, thereby eliminating the guesswork for others that follow in their path. Among the vendors involved, 3PAR has the most to gain as an emerging vendor by aligning with industry giants. HP stands to increase the reach of its blade server offerings through cross promotion. VMware has been notoriously playing Switzerland among its many solution partners at the expense of clear and actionable best practices; hopefully this announcement will signal a new wave of clarity from the server virtualization market leader.

By Andrew Reichman

Check out Andrew's research

Dell Buys iSCSI Storage Innovator EqualLogic

On Monday November 5th, Dell announced plans to buy iSCSI storage system vendor EqualLogic. This is a huge step forward for Dell, who until now has been content to sell relatively undifferentiated proprietary storage solutions based on server hardware and Microsoft operating systems, as well as OEM versions of industry leader EMC’s AX and CX lines. Fitting well with Dell’s theme of IT simplification, EqualLogic is an innovator, with an advanced technology that helps users take better advantage of IP-based storage, which is a lower cost alternative to Fibre Channel. EqualLogic’s design is cutting edge in three ways: 1) modular scalability that allows for easy capacity growth and increased controller power, cache, and I/O whenever capacity is added; 2) multiplexing architecture that allows multiple paths to each controller, effectively tripling the bandwidth of the 1 Gb Ethernet links; and 3) powerful software that automates many storage management functions, allowing companies without significant storage expertise to take advantage of the advanced storage functions included with every EqualLogic array.

This deal is interesting in a number of ways. For one, it is a validation of the iSCSI protocol which competes today for adoption with traditional Fibre Channel, and will compete with Fibre Channel over Ethernet (FCOE) going forward as 10Gb Ethernet becomes more prevalent. Also, this is a change of strategy for EqualLogic who had expressed the intention to go public in recent months. For a smaller emerging vendor to compete head-to-head with the giants of the storage industry in terms of R&D, interoperability and global sales and support is extremely difficult. Storage is an industry with very conservative buyers, high switching costs, and significant consideration of brand equity. Getting the deep pockets, manufacturing prowess, and brand recognition of Dell behind the innovation of EqualLogic is likely to be a powerful combination. There will be concerns around product roadmap continuity and focus, as well questions about placement of EqualLogic products within the current Dell storage lineup, but these are likely to be minimal in comparison to the strong technology benefits and complementary fit between the two companies.

By Andrew Reichman

Check out Andrew's research.