can expect to grow faster and more profitable the

use this technology as a tool for digital

evaluation.

The following content has been provided by the

University of Erlangen-Nürnberg.

The following content has been provided by the

University of Erlangen-Nürnberg.

Thank you very much and thank you Jurgen, for

inviting me.

Since I'm Vice Rector, I'm more used now to talk

about research policies and things like this.

I'm actually representing the lab, the laboratory I'm heading still,

which is called novel emerging computing system and technologies, which has inside areas

in terms of EDA and design of embedded system, high performance computing and security.

Here I will give you some ideas of what we are thinking strategically about the design

of autonomic heterogeneous computing architecture and some experiments we've done.

Obviously, the idea is much more generous strategy which is a long term five years vision

but we have done some parts of it, some pieces that we will try to put together in the next future.

So first, we can start with the context and the context basically is the fact

that we have an explosive growth and ubiquity of computation, communication and information

and there is a complexity, increasing complexity of heterogeneity of systems.

So current programming paradigms, methods,

management tools are basically inadequate to handle the scale, complexity,

dynamism and heterogeneity of the systems.

And so we need new methods to manage complexity and dynamic evolution.

So in one area which we didn't invent, I mean it's been there for more

than 15 years is autonomic computing.

Autonomic computing basically aims at allowing computing systems to be able to adapt

to the context by adapting their behavior and the resources, thousands time

of seconds in order to find the best way to try to accomplish a given set of goals

that despite the fact that the environment can change in terms of conditions and request.

So nature has evolved to cope with scale, complexity, heterogeneity, dynamism

and unpredictability and lack of guarantee by self-configuring,

self-adapting, self-optimizing, self-healing, self-protecting,

highly decentralized heterogeneous architecture that work.

So we try to get from this idea and try to find ways and tools that we leave the letter users

and the developers to specify and formulate to desire behavior rather

than fixed given solution to implement them.

So the pragmatic approach that we are taking is to try to separate knowledge,

policies and mechanism for adaptation.

Basically we are in a system in an area which is heterogeneous in terms of computing.

We have faster, smaller, cheaper, powerful, connected and heterogeneous computing systems

and one solution does not fit all.

As it was in the beginning of the computing area.

Both the design and the management of such system keeps increasing.

The behavioral complexity is also increasing and everything is interconnected

which requires different policies and function.

And in general, the behavior of this complex system

and mixed systems is too complex to predict.

So we start from the idea of ICT and biology.