Evaluation of PAMS' adaptive management services

The management of large-scale parallel and distributed applications is complex, due to factors such as centralized management architectures, lack of coordination and compatibility among heterogeneous network management systems, the dynamic characteristics of networks and application bandwidth requirements. The development of an integrated network management framework that is proactive, scalable and robust is a challenging problem. We present our approach to implement a Proactive Application Management System (PAMS). The architecture consists of two main modules: Application-Centric Management (ACM) and the Management Computing System (MCS). The ACM module provides the application developers with all the tools required to specify the appropriate management schemes to manage any QoS requirement or application attribute/functionality. The MCS provides the core management services to enable the efficient proactive management of a wide range of network applications. The services offered by the MCS are implemented using mobile agents. Furthermore, each MCS service can be implemented using several techniques that can be selected dynamically by invoking the corresponding mobile agent template for the service implementation. We present our preliminary results of evaluating PAMS management services to manage the performance and fault-tolerant execution of three applications of different sizes. The experimental results demonstrate that our agent-based approach can lead to significant gains in the performance and low-overhead fault management of parallel/distributed applications.