What people are saying - Write a review
We haven't found any reviews in the usual places.
Reliability of Individual Components
Alternative Approaches to Reliability Over a Long Mission
1 other sections not shown
16 clusters 32 clusters 5-year mission 64 clusters algorithm allel assumed bandwidth Byzantine resilience candidate architecture cessors Charles Stark Draper checkpoint cold spare processors communica communication links communication paths deadlock duplex redundancy errors estimate execution fail failure probability failure rate fault detection fault masking fault tolerance design fault tolerance features Fault Tolerant Parallel fault-tolerant system high number hypercube architectures initial configuration intercluster communication intercluster links interconnection intracluster Langley Research Center large number links needed long mission NASA NE's NUM[I number of components number of processors number of spare parallel computations PE's attached PE's connected PE's per cluster preliminary configuration probability of failure probability of system putations reconfiguration redundant computations reliability analysis reliability requirements rollback satellite SDI defense separate fault-containment regions SoHar sparing capability Strategic Defense Initiative subcomponents SURE program synchronization target application throughput tion Tolerant Parallel Processor topology transient faults V2-hour engagement mission Weibull distribution