Benefits of the Compaq Merger

Another direction to look in is the area of higher availability. In the course of Hewlett-Packard's merger with Compaq, it picked up some new and extensive high-availability technology. Again, given the reliance that enterprises today place on IT infrastructure, the usefulness of a system that simply does not go down at all is becoming more and more invaluable.

For example, the fault-tolerant systems from the former Tandem Corporation today function in environments where it is important to never lose a transaction. In a fault-tolerant environment, two computers run in parallel. If one fails, the other one picks up the load, backing up a little if needed to ensure that it collects any transactions that might have been dropped in the initial failure, and then continues on. This running in tandem was the main development that fueled the computer maker Tandem, which became part of Compaq, and hence part of Hewlett-Packard. The Intel Itanium processor family, with its machine-check architecture, is an ideal candidate for the next generation of these machines.

Today, we have software that actually watches our highly available systems and monitors sections or pages of memory where we may see degradations of performance. If you have a system that is running 24/7, there is no time to take the system down to run maintenance on the machine or its memory. In future Itanium-based systems, we will be able to take advantage of the architecture to improve these kinds of high availability and reliability techniques.

For server consolidation, the ideal is one giant server running complete software environments in virtual partitions, down to and including a single CPU—this can be made possible by the Intel Itanium processor family. The idea of computing as a utility becomes a closer reality with a centralized system that can handle multiple users, multiple networks, and multiple operating systems, consolidating all operations, helped by virtual partitions that can be made completely safe from software defects.

More impressively, the systems will be able to reallocate these partitioned resources as the demands change. Dynamic flexibility is the key here. To be able to handle different operating environments and also act as a fluid, not a rigid, system so that the most processing power gets put where it is needed is quite a challenge but offers an equally worthwhile benefit.