摘要:
An improved method and apparatus for time stamping events occurring on a large scale distributed network uses a local counter associated with each processor of the distributed network. Each counter resets at the same time globally so that all events are recorded with respect to a particular time. The counter is stopped when a critical event is detected. The events are masked or filtered in an online or offline fashion to eliminate non-critical events from triggering a collection by the system monitor or service/host processor. The masking can be done dynamically through the use of an event history logger. The central system may poll the remote processor periodically to receive the accurate counter value from the local counter and device control register. Remedial action can be taken when conditional probability calculations performed on the historical information indicate that a critical event is about to occur.
摘要:
An improved method and apparatus for time stamping events occurring on a large scale distributed network uses a local counter associated with each processor of the distributed network. Each counter resets at the same time globally so that all events are recorded with respect to a particular time. The counter is stopped when a critical event is detected. The events are masked or filtered in an online or offline fashion to eliminate non-critical events from triggering a collection by the system monitor or service/host processor. The masking can be done dynamically through the use of an event history logger. The central system may poll the remote processor periodically to receive the accurate counter value from the local counter and device control register. Remedial action can be taken when conditional probability calculations performed on the historical information indicate that a critical event is about to occur.
摘要:
A method and structure for a “dynamic CCR/sparse directory implementation,” includes maintaining state information of the main memory cached in the shared caches of the other compute nodes, organizing a cache directory so that the state information can be stored in a first area efficient CCR directory format, switching to a second sparse directory format if the entry is shared by more than one other compute node, and dynamically switching between formats so as to maximize the number of entries stored in the directory.
摘要:
A method and structure for an emulation system comprises of a plurality of field programmable gate arrays adapted to emulate nodes of a multi-node shared memory system, a plurality of cache directories, each connected to one of the arrays, and a plurality of global coherence directories, each connected to one of the arrays. Each of the global coherence directories maintain information on all memory lines remotely cached by each of the cache directories.