Autonomics for Grids and Datacenters (AGD'08)

Workshop in conjunction with the 8^th International Symposium on Cluster Computing and the Grid (CCGrid 2008)

The explosive growth of Grids, IT infrastructures and data centers have resulted in complex systems and applications whose control and timely management is rapidly exceeding human ability. Significant management challenges are resulting from, among other factors, their size, architecture complexity and distributed nature, their heterogeneity and the heterogeneity of their workloads, the large numbers of users and complexity and diversity of services provided, the need to provide strict RAS guarantees and adapt to unanticipated demands, the need to effectively manage power, etc.  These challenges, have resulted into systems are becoming increasing unreliable and brittle. Furthermore, current IT management solutions are costly, ineffective and labor intensive.

Autonomic computing has the potential to fundamentally address these challenges. Autonomic computing views IT infrastructures and their applications as closed loop control systems that need to be continuously monitored and analyzed, and plans corrective actions whenever any of the desired behavior properties or functionalities (e.g., performance, fault, security) are violated. From an intellectual perspective autonomics requires the integration and further development of techniques for monitoring, modeling, configuring, controlling and optimizing the behavior of resources and applications. Some techniques are inspired by strategies used by biological systems to deal with complexity, dynamism, heterogeneity and uncertainty.

Autonomic computing brings together a number of different areas such as artificial intelligence, self-organization/emergent behavior, distributed computing, virtualization, control, statistic, security, dependability and software engineering, with an overarching vision of integrated and adaptive automation, i.e., to effectively combine and simultaneously address requirements for performance, reliability, security, availability and cost-effectiveness in a holistic manner.

The goal of this workshop is to being together research focused on different aspects of autonomic computing as applied to Grid and data centers and their applications.   

PAPER SUBMISSIONS

Full papers (up to 8 pages in length) are invited on a wide variety of topics relating to autonomics for Grids and datacenters, as indicated above. Submissions will be evaluated on relevance, technical quality, and exposition. Papers must not have appeared before (or be pending) in a journal or conference with published proceedings, nor may they be under review or submitted to another forum during the AGD08 review process. Authors should submit full papers electronically (PDF) via email to agd2008@caip.rutgers.edu, and should use the IEEE formatting guidelines posted at the CCGrid website.

SPECIAL ISSUE

Authors of selected papers from the workshop will be invited to submit revised versions of their papers for a special issue of the Concurrency and Computation: Practice and Experience journal published by Wiley.