Daniel L. Palumbo and R. Lynn Graham, \Experimental validation of clock synchronization algorithms", NASA Technical Paper 2857, NASA Langley Research Center, Hampton, VA, July 1992.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....to be adjusted, but that the rest of the veri cation was una ected. Another change was stimulated by Dan Palumbo and Lynn Graham of NASA, who built equipment for experimenting with clock synchronization circuitry and found that the observed worst case skews were signi cantly better than predicted [30]. They showed informally that observation and theory could be brought into much closer agreement by changing the modeling of errors in the reading of clock values to use the fact that there is no read error when a processor reads its own clock. This improvement leads to reductions of 10 to 25 in ....

Daniel L. Palumbo and R. Lynn Graham, \Experimental validation of clock synchronization algorithms", NASA Technical Paper 2857, NASA Langley Research Center, Hampton, VA, July 1992.

....be adjusted, but that the rest of the verification was unaffected. Another change was stimulated by Dan Palumbo and Lynn Graham of NASA, who built equipment for experimenting with clock synchronization circuitry and found that the observed worst case skews were significantly better than predicted [30]. They showed informally that observation and theory could be brought into much closer agreement by changing the modeling of errors in the reading of clock values to use the fact that there is no read error when a processor reads its own clock. This improvement leads to reductions of 10 to 25 in ....

Daniel L. Palumbo and R. Lynn Graham, "Experimental validation of clock synchronization algorithms", NASA Technical Paper 2857, NASA Langley Research Center, Hampton, VA, July 1992.

....mechanicallychecked lemmas. It has been repeated just once: by Young [22] using the Boyer Moore prover. surements of a hardware implementation of the interactive convergence algorithm, that the observed worstcase skew is significantly better than that predicted by the standard Constraint C6 [13]. They showed that observation and theory could be brought into much closer agreement by using the fact that there is no error when a processor reads its own clock. The improved bound is easily incorporated into the formal verification by noting that the constraint in Lemma 4 is halved when ....

Daniel L. Palumbo and R. Lynn Graham. Experimental validation of clock synchronization algorithms. NASA Technical Paper 2857, NASA Langley Research Center, Hampton, VA, July 1992.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC