Jobs to be processed in distributed computing systems are characterized by a complex set of single interdependent tasks ("steps"). In the system, these steps represent the units of work associated with certain service requirements. Distributing the load among appropriate servers is a major issue in order to improve the system performance. "Traditional" load balancing approaches usually regard steps as independent and context-free. Scheduling approaches, on the other hand, assume complete knowledge of all steps to be scheduled and their job context, i.e. their interdependencies. However, the assumption of independent steps as well as the assumption of complete knowledge of the future workload situation are inadequate in many distributed systems. The real requirements are in the middle: Context-based workload predictions should be utilized as far as available, but it is mandatory to preserve the flexibility to react on unpredictable situations. This can be accomplished by context-sensit...

In Automonous Decentralized Systems (ADS), there is a high degree of uncertainty, especially in global state information and the performance and reliability of various resoufces. We model a distributed computer system of local and shared global servers, each of which can process jobs. The local servers insure a degree of autonomy and the global servers provide added compu-tational power and redundancy. When a job arrives, an agent decides whether to schedule the job locally, or whether to ship it to a global server. The agents have access to state information concerning current loads on the global servers and the response times of com-pleted jobs, but because of decentmlization this infor-mation is delayed, and perhaps lost, along communica-tion channels. The agents must make good decisions under the constraint that service rates are unknown and dynamic. We evaluate a deterministic algorithm and two randomizing algorithms and show that the de-gree of uncertainty determines which one is the most successful. 1