Sagnik Nandy

Learn More
Program profiling can help performance prediction and compiler optimization. This paper describes the initial work behind TFP, a new profiling strategy that can gather and verify a range of flow-specific information at runtime. While TFP can collect more refined information than block, edge or path profiling, it is only 5.75% slower than a very fast runtime(More)
A growing trend in the development and deployment of grid computing systems is decentralization. Decentralizing these systems helps make the more scalable and robust, but poses several challenges. In this paper we address one such problem - that of locating computing resources meeting specified requirements in a large scale heterogeneous system. The(More)
Overlapping communication with computation is a well-known technique to increase application performance. While it is commonly assumed that communication and computation can be overlapped at no cost, in reality they interfere with each other. In this paper we empirically evaluate the interference rate of communication on computation via measurements on a(More)
⋆ Abstract. This paper investigates the problem of autonomously allocating a large number of independent, equal sized tasks on a distributed heterogeneous grid-like platform, using only local information. We propose A-FAST (Autonomous Flow Approach to Scheduling Tasks), an efficient, scalable, dynamic and generic (imposing no restrictions on the topology)(More)
Pour chaque question, on attend une réponse de quelques lignesà une demi-page maximum. L'ensemble doit faire entre 2 et 4 pages. Question 1 : Quel est leprobì eme soulevé par cet article sur la modélisa-tion classique des communications (comme celle vue en cours) ? Question 2 : Supposons qu'un processeur A veuille envoyer des données aux processeurs B et C.(More)
The answer for each question should contain from a few lines up to half a page. The total should have between 2 and 4 pages. Question 1 : What is the problem mentioned in this article about the classical model for communications (as the one seen in class) ? Question 2 : Assume that a processor A has to send data to processors B and C. What is the(More)
Overlapping communication with computation is a well-known technique to increase application performance. While it is commonly assumed that communication and computation can be overlapped at no cost, in reality they interfere with each other. In this paper we empirically evaluate the interference rate of communication on computation via measurements on a(More)
Graduated with a distinction from BITS, Pilani. Top 0.5 percentile of more than 1 million high school graduates in India in 1997. My research interests lie in the areas of Distributed Computing, Performance Programming and Compilers. I strongly feel that we can make much better and more innovative use of the computing resources around us and am interested(More)
  • 1