Towards Autonomic Computing
The sheer density of blade servers is a strong argument for reducing the human input required for blade server management. Such systems must become more self-managing simply because the level of complexity is so great. Problems such as blade failures must not be allowed to bring down a complex subscriber-based service. In other words, the blade-based system must seamlessly self-heal with no operator intervention, perhaps failing over to backup blades with no loss of service.
An emerging area called autonomic computing (discussed in my next article) is an effort to address this class of problem. We can expect that as autonomic computing models are deployed, the role of the data center manager will move up the value chain.