Pankaj Khanchandani

Learn More
The competitive analysis of an algorithm Alg consists of two separate steps. First, we show that for an arbitrary problem instance, the result of Alg is asymptotically at most a factor r worse than the the optimal offline result. This yields an upper bound on Alg’s result, that is costAlg ≤ r · costOpt + c. If the task is to show that Alg is(More)
Figure 1: Original version In its most primitive form, coined original version from here on, it partitions a dataset into 2 (where R ≤ 32) partitions as shown in Figure 1: in the first pass over the data, we count for each partition the number of entries that will be sent to it (lines 1-2). From this generated histogram, we calculate the start index of each(More)
We revisit the approach to Byzantine fault-tolerant clock synchronization based on approximate agreement introduced by Lynch and Welch. Our contribution is threefold: • We provide a slightly refined variant of the algorithm yielding improved bounds on the skew that can be achieved and the sustainable frequency offsets. • We show how to extend the technique(More)
  • 1