Storage

08:33 AM
David Hill
David Hill
Commentary
50%
50%

Crossroads' StrongBox: Securing the Future of Data Archiving

If you were asked which storage product--hard disk drives (HDDs) or tape--would decline precipitously in market share during the next five to 10 years, you would probably be one of the vast majority of people who said tape. You would also be wrong. Winning storage technologies have to excel on at least one dimension, either performance or capacity, and tape excels on capacity.

If you were asked which storage product--hard disk drives (HDDs) or tape--would decline precipitously in market share during the next five to 10 years, you would probably be one of the vast majority of people who said tape. You would also be wrong. Winning storage technologies have to excel on at least one dimension (either performance or capacity) and tape excels on capacity.

As a result, as active archiving becomes more important to your enterprise, tape will play a major role to meet the high-capacity requirements that the active archiving IT revolution will demand. A number of IT vendors, both large and small, will play significant roles in that revolution. However, Crossroads and its StrongBox product seem likely to play a crucial part in tape’s upcoming significant position in active archiving.

Before we examine Crossroads’ StrongBox, we need to understand the relative future roles of HDDs, solid state drives (SSDs) and tape. The successor technologies to flash memory (which are already on the drawing boards in storage R&D labs) will dominate the performance segment of the storage market. HDDs simply are not increasing in speed fast enough to overcome the advantages of solid state device technologies. Although HDDs may (or may not) retain a relative price advantage over solid state devices, the absolute cost (that is, the money that enterprises have to shell out) will fail to overcome the value benefits of the solid state devices when high performance (as measured in response time latency requirements) is mandatory.

What about capacity? Well, we have already seen data migrate from high performance disks (say, Fibre Channel disks) to more friendly cost-per-gigabyte drives (such as SATA disks). Why isn’t this sufficient? The answer is that although some might disagree that tape, at sufficient scale, is much more cost efficient than disk, the majority of data is fixed content data (that is, not subject to change), which does not require fast response times, so the most cost-efficient technology wins (especially when the difference in efficiency is significant).

But, wait, some might say. What about manageability? What this boils down to is that disk can use a file system for management and, until recently, tape did not have that advantage. But with the introduction of the linear tape file system (LTFS, a single, non-proprietary media storage standard for tape), that advantage has gone by the boards. Tape can now be managed as if it were disk.

But, wait again, say the critics. Huge tape cartridges still have to store files in a linear fashion, and finding a particular file may take time--say, a minute or more compared with the seconds required by HDDs. Note, however, that the first part of each file can be placed on a "disk" cache, where disk may be either an HDD or SDD device. This may be enough to save precious time.

Moreover, recall that information is typically stored on tape for long-term retention, which means that access is very infrequent and search time may not be much of a factor. If a piece of information becomes hot in the sense that it suddenly becomes active (that is, frequent accesses), such as biographical files about a famous person who just died, then they can be positioned on disk until the hot spell subsides. Also, some data retrieval can be scheduled, such as medical images called up prior to a doctor’s appointment, and other files can be subject to analytical processing (such as big data), where the initial latency is so small as to be irrelevant to the overall length of processing.

David Hill is principal of Mesabi Group LLC, which focuses on helping organizations make complex IT infrastructure decisions simpler and easier to understand. He is the author of the book "Data Protection: Governance, Risk Management, and Compliance." View Full Bio
Previous
1 of 3
Next
Comment  | 
Print  | 
More Insights
Hot Topics
5
Do We Need 25 GbE & 50 GbE?
Jim O'Reilly, Consultant,  7/18/2014
White Papers
Register for Network Computing Newsletters
Cartoon
Current Issue
Video
Slideshows
Twitter Feed