Search

Meenaradchagan Vishnu

from San Jose, CA
Age ~58

Meenaradchagan Vishnu Phones & Addresses

  • 6377 Canterbury Ct, San Jose, CA 95129 (408) 257-2551
  • Petaluma, CA
  • 117 Lexington Dr, Cranberry Twp, PA 16066 (724) 772-6683
  • Cranberry Township, PA
  • Cotati, CA
  • Santa Clara, CA
  • Cupertino, CA
  • 6377 Canterbury Ct, San Jose, CA 95129

Work

Position: Clerical/White Collar

Education

Degree: High school graduate or higher

Publications

Us Patents

Connection Rearrangement In Communication Switches

View page
US Patent:
7187672, Mar 6, 2007
Filed:
May 15, 2002
Appl. No.:
10/147446
Inventors:
Meenaradchagan Vishnu - Petaluma CA, US
Assignee:
Calix Networks, Inc. - Petaluma CA
International Classification:
H04Q 11/00
H04Q 3/00
H04J 3/02
US Classification:
370360, 370376, 370386, 370539, 340 228
Abstract:
A processor is programmed to reduce a problem of adding a new connection to a time-space-time (TST) switch of a communication network into a problem of graph theory, and to solve the problem using a heuristic instead of an exact algorithm. A solution, if provided by the heuristic, is used to rearrange the connections in the TST switch. Several embodiments of such a programmed processor reduce a connection rearrangement problem of a TST switch into any one of the NP-complete problems (such as the vertex coloring problem or the boolean satisfiability (SAT) problem). In some such embodiments, the processor is programmed based on the Brélaz heuristic to find a solution to the vertex coloring problem. In other embodiments, other heuristics, such as a genetic algorithm, may be used.

Arbiter For An Input Buffered Communication Switch

View page
US Patent:
20040017804, Jan 29, 2004
Filed:
Jul 19, 2002
Appl. No.:
10/199996
Inventors:
Meenaradchagan Vishnu - Petaluma CA, US
International Classification:
G06F013/00
US Classification:
370/386000, 370/412000
Abstract:
An arbiter for a switch maintains a pair of counters for each flow of traffic at each input port: one counter (also called “first counter”) to indicate an ideal transfer of traffic, and another counter (also called “second counter”) to indicate the actual transfer of traffic. Both counters are incremented when traffic is received by the input port, and the second counter is decremented when a unit of traffic (such as a cell or packet) is about to be transmitted whereas the first counter is decremented in a fractional manner (relative to the unit of traffic) in each period of arbitration, based on available bandwidth. The arbiter selects one of the output ports (also called “winning output port”) of the switch, based at least partially on values of the two counters for each flow from the input port to one of the output ports, and generates a signal to approve transfer of traffic from the input port to the winning output port. In several embodiments, the above-described flow can be for either high priority traffic or for low priority traffic, and any bandwidth leftover from transferring high priority traffic is used in transferring low priority traffic. Specifically, the arbiter maintains additional counters for each port indicative of total bandwidth being used, and the additional counters are used to allocate leftover bandwidth in an iterative manner, until a flow from an input port to an output port is saturated, at which time the saturated flow is removed from iteration.

System, Method, And Computer Program Product For Implementing Software-Based Scoreboarding

View page
US Patent:
20150220341, Aug 6, 2015
Filed:
Feb 3, 2014
Appl. No.:
14/171671
Inventors:
- Santa Clara CA, US
Michael Alan Fetterman - Boxborough MA, US
Olivier Giroux - Santa Clara CA, US
Jack H. Choquette - Palo Alto CA, US
Xiaogang Qiu - San Jose CA, US
Shirish Gadre - Fremont CA, US
Meenaradchagan Vishnu - San Jose CA, US
Assignee:
NVIDIA Corporation - Santa Clara CA
International Classification:
G06F 9/30
Abstract:
A system, method, and computer program product are provided for implementing a software-based scoreboarding mechanism. The method includes the steps of receiving a dependency barrier instruction that includes an immediate value and an identifier corresponding to a first register and, based on a comparison of the immediate value to the value stored in the first register, dispatching a subsequent instruction to at least a first processing unit of two or more processing units.
Meenaradchagan Vishnu from San Jose, CA, age ~58 Get Report