Other Industry Trends in High Availability

Third-party hardware innovations continue to lead the way in evolving high availability. Microsoft will not go into the hardware business (I don't think), so look for most of the major hardware vendors to be pushing heavily into “blade” and other “hot swappable” hardware architectures for years to come. This will also be much more tightly tied to the virtual machines approach (and the virtual server) so successfully utilized in the Unix world.

Disk/media vendors are quickly gobbling up horizontally compatible companies and combining their expertise to provide more reliable and available content platforms. An example of this was when EMC recently bought Documentum and then rolled out a completely integrated fault-tolerant and highly available document management system.

Disaster recovery is rapidly evolving to become an outside service rather than an in-house capability. Within the coming years, there will be numerous subscription models for any size company to subscribe to that can be used in the event of a disaster. This will include remote and secure distribution of a company's data and applications that can be switched on in a moment's notice.

General pods (or grids) of computing will slowly take over many organizations' data processing needs. IBM is pushing “on demand” centers for companies that need extended virtual processing capabilities in a moment's notice. Intel is building server farms around the world to position itself to accommodate a global move to outsourced computing, and many other companies are building up variations on this theme (such as Oracle and HP). And, most Application Service Providers (ASP) vendors simply list high availability as the minimum capability that they offer.