Later this week we will announce a pretty comprehensive set of services and software to help businesses develop increasingly autonomic, i.e., self-managing IT infrastructures.
Autonomic computing is an initiative we started at IBM in 2001 when we recognized that the same very positive forces that are driving the growth of the IT industry could be responsible for its biggest problems, namely the almost exponential growth in complexity and the rising costs of managing this complexity.
What are these forces that are simultaneously driving growth as well as complexity? I think of three main ones. First, information technologies have been getting much more powerful and inexpensive at a prodigious rate, and are giving us larger IT infrastructures while simultaneously becoming ubiquitous, as computing is increasingly embedded all around us. Second, is the good old Internet. Just about all of the elements of the IT infrastructure as well as the growing number of devices are connected via the Internet forming a gigantic distributed system letting us access anything from anywhere. And finally, as IT is increasingly integrated into all aspects of business, society and our personal lives, the stuff has to work. We are as dependent on IT as we are on other essential infrastructure like electricity and transportation. I was going to add communications to this list except that with Voice over IP the IT infrastructure is already becoming the telecommunications infrastructure.
Not surprisingly, when you contemplate a world with billions of systems and trillions of pervasive devices all connected via the Internet, and millions of businesses and billions of people dependent on that infrastructure, you have a big time grand challenge, perhaps the grandest of all challenges in the IT industry: how to attack this growing complexity by making computing systems increasingly self-managing. So, in 2001 we issued a challenge not just to ourselves but to the whole industry to address this problem by making sure that we make autonomic computing a major part of all products and services we design, as well as investing in R&D to solve the very difficult technical problems in the area.
We have been taking this challenge really seriously in IBM, especially given our focus on business and infrastructure solutions. Although we did not use the term, autonomic computing has been part of the mainframe heritage in IBM, since the essence of their success is the fact that mainframes work just about all the time. But, as new modes of computing arose with the advent of PCs, Client/Server, the Internet and Pervasive Computing, and as competitive pressures forced us all to get products to market faster at lower prices, the IT industry, including IBM, got sloppy. Since we announced the initiative in 2001, we have been building in autonomic capabilities into just about everything we do, and announcing new products and services like those we just announced this week. We formed an Autonomic Computing unit to oversee all these efforts across the company, and significantly increased our research focus on autonomic computing, both within IBM and in our work with universities.
We named the autonomic computing initiative after the autonomic nervous system, which is the portion of every biological system, including us humans, whose job is to keep us alive and evolving into the future, without having to consciously think about it and take action. I think we have made great strides in improving the autonomic capabilities of our computer systems, but in comparison with biological systems, our computers are still in the very early stages of what needs to be done. Autonomic computing, and the overall management of complexity will continue to be one of the most important challenges in the IT industry for a long time to come.
Comments