My other column gave a rationale for why the mixture of grid, utility computing and Information Lifecycle Management is going to drive a lot of buying decision-making in 2005.
But what sort of products will be providing this technology mix? First, keep in mind that we are talking about increasingly complex systems, and that the complexities of those should be presented to the human management team in as simplified a way as technology allows. Next, also remember that the only efficient way to implement technologies like this is to automate them to the highest possible degree.
Utility computing will mean distributed environments, and distributed environments mean multi-vendor, multi-platform, multi-domain heterogeneity. In support of this requirement, expect adherence to industry standards to become mandatory in just about all enterprise RFIs by year-end.
We should expect to see increased strength in the sales of storage resource management (SRM) and data migration software in high-end computing facilities.
Furthermore, the importance of automated provisioning and data movement will be paramount at the sort of sites we are talking about; anyone expecting to participate in on-demand environments should expect to invest in high-value (read "not cheap") software.
In connection with this point: as "value" will have much more to do with operational expense than anything else, IT buyers will want to check out both the amount of automation and the quality of the automation that their vendors offer. Complexity underneath and ease of use in the humanware interface should be the mantra here.
No matter how good the automation is however, it can't provide optimal performance unless storage managers have information about more than just storage. Utility computing (and grid) requires not only an ability to manage individual parts of the system, but also the ability to manage the connectivity between the various components that comprise the system.
Connectivity will be enabled by standards-adherence. But when things go wrong nothing will take the place of really sharp analytics that can look across several domains (storage and networking for example, or storage and servers), monitor the events that take place (and perhaps even anticipate them), and then present the human managers with accurate analysis and guidance regarding systems events that may span the entire IT system. This will eliminate hours of downtime troubleshooting and allow managers to focus on actually fixing problems rather than just sorting through a problem's many symptoms.
Two important developments will drive much of what happens with storage hardware. SATA-based systems will find increased acceptability within enterprise IT- and not only in lower-tiered applications. Then, about mid-year, expect that serial-attached SCSI (SAS) systems will start showing up to provide even greater functionality.
The result of all this will be that buyers will start playing mix-and-match with their disk technology, crafting systems that match their requirements much more closely than has ever been previously possible.
Finally, keep your eyes open for improved solutions for remote-office storage management problems. These will appear in a variety of formats - including (but not limited to) service providers and wide area file services. Significantly, several of these solutions will have applicability in the small-and-midsize business segment as well.
And the Yankees will fail to win the American League again, no matter how much they have invested.
We will take a look at some new IT solutions next week.