What the Future Holds

 < Day Day Up > 



What does the future hold for hard drives? In addition to major improvements in monitoring tools, we can look forward to even greater storage capacity. Although several exciting new technologies (e.g., using DNA molecules) can greatly boost capacity, traditional magnetic disks will be with us in the enterprise for a long time. So, yes, disks will probably eventually disappear, but it will not be for many years to come, and even then they probably will go on to live far longer than predicted, much in the same way that tape drives are still very much alive and well, over a decade after the industry sounded their death knell.

Similarly, in the area of data preservation, RAID will continue to gain popularity. Until recently its price tag has limited its application to the more expensive and critical servers. That appears to be changing, with RAID now being standard in most mid-sized machines and even appearing more often in lower-end models.

A greater shift toward centralized storage will occur at the enterprise level. Although 92 percent of the market is currently using direct attached storage, Forrester predicts that by 2003 the use of both Network Attached Storage (NAS) and Storage Area Networks (SAN) will double. At the other end of the spectrum is a growing need for portable disks, such as those used in digital cameras and MP3 players.

One major void still has to be filled, though: heterogeneous disk monitoring tools that can be run across a network. One vendor recently made a valiant, but unsuccessful, attempt to provide this service. It ran afoul of the proprietary nature of RAID arrays and disk controllers. Most manufacturers do a fine job of providing tools that help users run that vendor's own disks; however, these tools rarely do very well with other manufacturers' products. Yet, what organization has only one make of disk throughout the enterprise? As a result, IT managers are forced to go from console to console to read alerting lights and event logs in multiple management systems, not to mention going to various physical locations to check on disk health.

The ideal would be one screen that operates irrespective of disk type or model and that tells users everything they need to know about their disks — their performance (using many parameters such as throughput and error rates), speed, indications of imminent failure, free space, quota information, and more. The SMART standard built into many current disks was a start, but it has too many limitations and was implemented in much too proprietary a fashion to be really useful.

Compaq Insight Manager (CIM) took things a step further. The latest version allows users to monitor drives, CPUs, and other devices on Compaq as well as some other vendor machines. It is hoped that Hewlett-Packard will retain it and expand it, but even CIM left us in the hole when it came to white box servers, and it was not that great with non-Compaq machines. I expect someone will step forward to fill the gap, providing a method that either bypasses vendor proprietary controls or unites the vendors in acceptance of a standard and opening up their architectures (a long shot). Any takers? When, or if, this utility ever arrives, it will offer what IT needs more than ever — a way to easily manage and monitor all the drives and a means of predicting usage and preventing disastrous failures. With that in hand, disk management would become much simpler for all concerned.



 < Day Day Up > 



Server Disk Management in a Windows Enviornment
Server Disk Management in a Windows Enviornment
ISBN: N/A
EAN: N/A
Year: 2003
Pages: 197

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net