Storage checkpoint 2005: a new chapter.
The rate of change has slowed from the pre-2001 period. Venture capital money and new business ideas are more slowly entering the system, further slowing innovation. The year 1998 saw 27 new storage companies launched. In 2000, 88 storage companies were launched with a first round of funding. Today, only business plans that solve real problems are being funded. As a result, just four new storage companies were funded in 2004, suggesting that the rate of innovation has slowed. In 2001, the number of storage firms reporting net losses hit 69%. Vendor roadmaps have been pushed out as storage vendors derive lower profitability and, therefore, have less money to invest in their future development activities. The trend for profitability shifted and headed in the right direction as 46% reported losses in 2003. A record for the past three years, 79% of the companies reporting financials experienced revenue growth in 2003 over 2002 offering more encouragement.
Beginning in 2004, a new game with new rules was clearly under way. These rules described a new value system for IT where, for the first time, the overall value proposition became more important than raw price. The impact of this transformation on the storage industry will have lasting effects most likely altering much of the traditional thinking and defining a need to navigate through the next era now awaiting the storage industry.
2005 ... and Beyond
With high availability on the top of the list for every IT-oriented business, today's goal to achieve the mainframe (the z/Series) level of five 9s system availability and 100% security levels is attempting to spread beyond the mainframe to the Unix, Linux and even more importantly, the Windows platforms. This level of availability most likely will never be attained by these platforms, as their inherent limiting architectural design issues are significant. The mainframe is slowly declining in the number of installations and machines installed worldwide but MIPS consumption (the measure of a mainframe's computational power) are at an all time high. Nonetheless, the high availability, I/O and storage management capabilities of the mainframe remain the ultimate goals for non-mainframe systems.
Mandatory Data protection
For years, IT managers thought about data security in terms of protecting data from disk crashes or tape malfunctions. Today, storage security requirements have moved well beyond the storage subsystem and local network and into large geographically spread enterprises. Under mounting threats from cyber-terrorists, countless compliance regulations are now legally mandating that most businesses worldwide secure data at all levels of the local and remote enterprise. Securing data now includes protecting data from hardware, software, intrusion (WORM, virus, hackers) and theft. Data protection from theft has given momentum to encrypting selected data at rest and several providers are delivering encryption capabilities.
As devices become increasingly more reliable, intrusion becomes an ever-larger threat to attaining high levels of IT availability and data security. Driven by cyber-terrorists, hackers and other malicious types, intrusion and a growing number of data security issues mandate that companies immediately implement dramatic improvements to their current and very porous IT security systems. It's likely that the savvy disk, tape and optical storage vendors will offer a wide range of storage security functionality including features emphasizing WORM capability and encryption appliances for stored data. These new systems will use hyper-firewalls, advanced encryption and eventually biometrics to enable the entire data protection market to grow from $17 billion in 2001 to more than $40 billion by 2006 making it larger than the disk, tape and optical storage hardware industry combined. Intrusion systems must soon reach bulletproof status. Protecting data throughout its lifetime is a piece of the popular ILM messaging. Getting all the way there, however, will be a difficult and over-arching goal for the storage solution providers.
Storage management looks for a home
The implementation and ongoing maintenance of as many as three different storage topologies, SAN, NAS and DAS, has proven to be a costly venture for most businesses. Storage management capability now exists in the server, the fabric and the storage subsystems creating many choices surrounding where this function should reside creates an overlay of confusion. SANs have appropriately generated widespread appeal, but installations have occurred more slowly than expected, as the implementation process remains resource-intensive at a time when resources are shrinking. A recent study of Fortune 1000 storage professionals indicated that SAN and NAS demand are both expected to grow at 51% for 2005. This remains a healthy growth rate but below the previous 70% growth predictions. During 2005, the storage network/fabric will begin to take over some of the traditional storage management load. Expect the storage fabric or the intelligent network switch to provide decision capability regarding where data is stored and how it is protected. Leading switch vendors all have development efforts underway to add storage management intelligence to their switch platforms during 2005 and beyond though market acceptance remains slower than hoped for.
Scalability: More than capacity
Historically, storage vendors have described scalability in terms of capacity increases and maximum limits but, as we have witnessed in the past year, if you don't scale performance and capacity at the same rate, throughput bottlenecks result and the overall capability of the subsystem decreases. Savvy businesses now recognize that scalability involves more than just increasing capacity and storage subsystem vendors are finally beginning to accept that scalability requires the two dimensions of capacity and performance. Later, connectivity will join capacity and performance in the scalability equation as subsystem-level throughput and availability are needed to enable the storage infrastructure to maintain equilibrium.
The value of IT has shifted
Many users today still look at the hardware purchase price as their primary purchase criteria. This approach has become increasingly unfortunate and reflects a legacy viewpoint that the value of the IT infrastructure exists in hardware. This is like measuring the value of the television industry by the number of sets sold (the old rules) rather than the value of the content being transmitted by television (the new rules). With hardware prices falling 30-40% per gigabyte annually, clearer thinking in the storage industry is shifting to the value of the data and away from the devices that store it. Digital information is growing at 30 to 40 percent annually. 2005 is the year that businesses are seriously developing a tiered storage model based on the value of information over its lifetime and then maps it to the storage system that best meets those requirements.
As the Unix, Linux and Windows storage systems (inappropriately called open systems) have become increasingly larger and now account for more than 85% of installed storage capacity, a defacto standard HSM software system for these systems is desperately needed to control disk growth and is long overdue. For years, HSM has been a widely accepted and standard practice on mainframe computers, providing significant financial and management value. The year 2005 should identify a legitimate HSM cross-platform software product that resembles and provides some of the powerful mainframe-like storage management functions. A solid HSM simplifies the management of tiered storage.
The storage industry has embarked on a new course leaving behind many of the tenants that defined the way the storage business was conducted for the last 30 years. Expect cutting-edge technologies that have the potential to offset the losses in the old markets to create new markets and possibly industries. New technologies rarely change the world overnight and predicting their inflection point is difficult. Nonetheless, expect to be looking directly at transparent data movement, libraries of disks, non-volatile RAM chips, object-based storage, miniaturized nano-technologies and atomic-level manipulation as potential breakthroughs in the years ahead. Companies and individuals in most of the world rely on information technology to the point that their survival is threatened if it is not available. Finally, it is not an option to simply react to the changing and unfamiliar landscape; setting a course for future is your best bet to navigate successfully through the challenges ahead. The events of the past have signaled that the IT industry of the future will be different than the IT industry of the past. With the economic decline of the past three years now subsiding, look for the year 2005 to launch new storage initiatives that identify a pathway to higher availability and someday near bulletproof security. Many of these offerings will have little or no resemblance to their predecessors. Spend a little time looking beyond the daily fires or the 90-day earnings window; it's time to prepare for your future.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Storage Networking|
|Publication:||Computer Technology Review|
|Article Type:||Industry Overview|
|Date:||Mar 1, 2005|
|Previous Article:||Tape automation extends its 5-year run: advantages keep tape center stage.|
|Next Article:||Tape is here to stay.|
|BACKUP WORKING GROUP.|
|Applications For Shared Data Clusters.|
|Network Administrators Survival Guide.|