Kedar Dhamdhere

Finding (recently) frequent items in distributed data streams

We consider the problem of maintaining frequency counts for items occurring frequently in the union of multiple distributed data streams. Naive methods of combining approximate frequency counts from multiple nodes tend to result in excessively large data structures that are costly to transfer among nodes. To minimize communication requirements, the degree of precision maintained by each node while counting item frequencies must be managed carefully. We introduce the concept of a precision gradient for managing precision when nodes are arranged in a hierarchical communication structure. We then study the optimization problem of how to set the precision gradient so as to minimize communication, and provide optimal solutions that minimize worst-case communication load over all possible inputs. We then introduce a variant designed to perform well in practice, with input data that does not conform to worst-case characteristics. We verify the effectiveness of our approach empirically using real-world data, and show that our methods incur substantially less communication than naive approaches while providing the same error guarantees on answers.

Metric embeddings with relaxed guarantees

We consider the problem of embedding finite metrics with slack: we seek to produce embeddings with small dimension and distortion while allowing a (small) constant fraction of all distances to be arbitrarily distorted. This definition is motivated by recent research in the networking community, which achieved striking empirical success at embedding Internet latencies with low distortion into low-dimensional Euclidean space, provided that some small slack is allowed. Answering an open question of Kleinberg, Slivkins, and Wexler (2004), we show that provable guarantees of this type can in fact be achieved in general: any finite metric can be embedded, with constant slack and constant distortion, into constant-dimensional Euclidean space. We then show that there exist stronger embeddings into /spl lscr//sub 1/ which exhibit gracefully degrading distortion: these is a single embedding into /spl lscr//sub 1/ that achieves distortion at most O(log 1//spl epsi/) on all but at most an /spl epsi/ fraction of distances, simultaneously for all /spl epsi/ > 0. We extend this with distortion O(log 1//spl epsi/)/sup 1/p/ to maps into general /spl lscr//sub p/, p /spl ges/ 1 for several classes of metrics, including those with bounded doubling dimension and those arising from the shortest-path metric of a graph with an excluded minor. Finally, we show that many of our constructions are tight, and give a general technique to obtain lower bounds for /spl epsi/-slack embeddings from lower bounds for low-distortion embeddings.

On two-stage stochastic minimum spanning trees

We consider the undirected minimum spanning tree problem in a stochastic optimization setting. For the two-stage stochastic optimization formulation with finite scenarios, a simple iterative randomized rounding method on a natural LP formulation of the problem yields a nearly best-possible approximation algorithm. We then consider the Stochastic minimum spanning tree problem in a more general black-box model and show that even under the assumptions of bounded inflation the problem remains log n-hard to approximate unless P = NP; where n is the size of graph. We also give approximation algorithm matching the lower bound up to a constant factor. Finally, we consider a slightly different cost model where the second stage costs are independent random variables uniformly distributed between [0,1]. We show that a simple thresholding heuristic has cost bounded by the optimal cost plus

Non-Clairvoyant Scheduling for Minimizing Mean Slowdown

\frac{\zeta(3)}{4}+o(1)

Scheduling for flow-time with admission control

.

Algorithms for Efficient Near-Perfect Phylogenetic Tree Reconstruction in Theory and Practice

AbstractWe consider the problem of scheduling dynamically arriving jobs in a non-clairvoyant setting, that is, when the size of a job in remains unknown until the job finishes execution. Our focus is on minimizing the mean slowdown, where the slowdown (also known as stretch) of a job is defined as the ratio of the flow time to the size of the job. We use resource augmentation in terms of allowing a faster processor to the online algorithm to make up for its lack of knowledge of job sizes. Our main result is that the Shortest Elapsed Time First (SETF) algorithm, a close variant of which is used in the Windows NT and Unix operating system scheduling policies, is a

Fixed parameter tractability of binary near-perfect phylogenetic tree reconstruction

(1+\epsilon)

Simple reconstruction of binary near-perfect phylogenetic trees

-speed,

Metric Embeddings with Relaxed Guarantees

O((1/\epsilon)^5 \log^2 B)

Approximation Algorithms for Minimizing Average Distortion

-competitive algorithm for minimizing mean slowdown non-clairvoyantly, when