Abstract: Failures are detected in servers of a transaction processing system, and transactions are routed to less failure prone servers in the system. Servers in the transaction processing system which are faulty for some transaction classes but good for others are detected, and such servers are used in a judicious manner to maximize the throughput and minimize the response time of the system. Error prone servers are occasionally probed to determine if they have improved in terms of their error characteristics. The mechanism implemented consists of three elements. The first is the selection of a routing algorithm based on the state of the transaction processing system. Second, transactions are used to probe systems considered too faulty for use in order to determine if they have improved in terms of their failure characteristics. Finally, soft ABENDs are detected.

Abstract: A packet switch system having a buffer occupancy reduction mechanism for controlling data flow through switched nodes of the system to avoid congestion and reduce required buffer storage at the nodes. For an isochronous connection, buffers are initially allocated at each switching node connection to ensure listless transmission of packets. A buffer occupancy trace for the connection is recorded and the delay time of an isochronous packet at a particular switch port is returned to a preceding switch port. The preceding switch port employs the feedback message to delay subsequent packets through the connection to reduce the queuing time at the particular switch port. Once the isochronous connection has stabilized, buffer reallocation is performed wherein a scheduler at each switch node along the connection attempts to combine buffer allocations for different isochronous connections. This occurs provided the corresponding buffer occupancy traces of the isochronous connections do not overlap.