Fault-tolerant techniques that can cope with system failures in software distributed shared memory (SDSM) are essential for creating productive and highly available parallel compu...
Suppose a long-running analytical query is executing on a database server and has been allocated a large amount of physical memory. A high-priority task comes in and we need to ru...
Badrish Chandramouli, Christopher N. Bond, Shivnat...
— Large Clusters, high availability clusters and Grid deployments often suffer from network, node or operating system faults and thus require the use of fault tolerant programmin...
In this paper we investigate the effect of three time-triggered system rejuvenation policies on service availability using a queuing model. The model is formulated as an extended ...
Interconnect tuning and repeater insertion are necessary to optimize interconnectdelay, signalperformanceandintegrity, andinterconnectmanufacturability and reliability. Repeater i...