Enabling autonomic computing with computational intelligence

阻碍计算机软件系统未来发展的一个障碍是它们日益增长的复杂性。在不久的将来，我们将达到这样一种状态，即这些系统的操作对人类专家来说将是棘手的。在自主计算范式中提出了一种解决方案，其目的是开发由许多相互作用的自我管理模块组成的计算机软件系统。这可以部分地通过使用用于服务发现和模块之间通信的开放标准来实现。但自主计算也需要在软件系统中注入智能和健壮性，以便将人类操作员从低级细节中解放出来。计算智能包含许多结合学习、适应、进化和模糊性元素的技术。使用这些技术来开发自我管理软件系统看起来很有希望，并且将在当前研究计划的背景下进行探索。

An obstacle to future developments of computer software systems lies in their ever-increasing complexity. In a not-so-distant future, we will reach a state where the operation of these systems will be intractable for human experts. A solution to this has been proposed in the autonomic computing paradigm, which aims at developing computer software systems made of many interacting self-managing modules. This can be achieved in part through the use of open standards for service discovery and communication between modules. But autonomic computing also requires the injection of intelligence and robustness in the software systems, in order to free human operators from low-level details. Computational intelligence encompasses many techniques combining elements of learning, adaptation, evolution, and fuzziness. Using these techniques for the development of self-managing software systems looks promising, and will be explored within the context of the current research program.