An Evaluation of Deadlock-Handling Strategies in Semiconductor Cluster Tools

Abstract

We study the deadlock handling performance of a single-blade cluster tool and present a comparison of average flowtimes under deadlock detection and resolution versus prevention. Two detection and resolution policies are tested: either resolve permanent deadlocks or resolve both permanent and transient deadlocks. A permanent deadlock requires external intervention to resolve the deadlock, whereas a transient deadlock has a positive probability that the deadlock will resolve itself over time. Prevention averts deadlock by providing sufficient in-process buffer spaces. Our experiments indicate interplay of process and robot transfer times dictate the choice of deadlock strategy. Under low robot transfer times relative to process times, providing sufficient in-process buffer to prevent deadlock or resolving both permanent and transient deadlocks can be equally effective. We conclude with some practical guidelines for operating and designing cluster tools under deadlock conditions. Index Terms-Cluster tool design, deadlock detection and resolution, deadlock handling, deadlock prevention, flexible manufacturing systems, real-time control of cluster tools.

Deadlock Prevention Algorithm in Grid Environment

“…Therefore it would be more efficient if we look for an efficient deadlock prevention mechanism for distributed real time systems. The deadlock avoidance strategies make system performance to suffer [7], although deadlock detection may be effective, but it costs a lot in business transaction services [8] and detection of false deadlocks leads to wastage of resources. However deadlock resolution mechanisms are there but many real time applications are not well suited for run time deadlock resolution [9].…”

Scalable Coordination Mechanism to Maintain Throughput of Dynamic Multiagent Networks

“…Deadlocks can be of two types, permanent and transient deadlock. A permanent deadlock cannot be resolved without external intervention, whereas a transient deadlock resolves itself over time (Venkatesh & Smith 2005). The probability of having deadlocks increases when capacity of buffers in system is reduced.…”

“…As permanent bottlenecks bring the system to standstill, issue of bottlenecks has to be addressed in order to have an autonomous system that can work with limited buffer capacities. Because of the complexity of system, avoidance, identification and resolution of deadlocks in semiconductor manufacturing processes is a difficult problem and various sophisticated techniques are being investigated under current research (Venkatesh & Smith 2005). As the techniques for managing deadlocks in semiconductor fabrication processes are still under investigation, in order to focus our attention on behavior of CABS, we have developed an distributed algorithm that avoids permanent deadlocks in system.…”