International Business Machines Corporation

Larry Joseph Stockmeyer

Jaishankar Moothedath Menon

Randal Chilton Burns

Ronald Fagin

Miklos Ajtai

Ching-Tien (Howard) Ho

The present invention provides a method and system of interconnecting L processors of a parallel computer to facilitate torus partitioning, (a) where each of the processors includes a processing unit and a switch, (b) where the switch includes a first external port, a second external port, a third external port, a fourth external port, a first internal port, and a second internal port, (c) where the L processors comprise R non-overlapping partitions, (d) where each of the partitions comprises the processing unit of at least one of the processors, and (e) where L is an integer &#x2267;2 and R is an integer &#x2267;1. In an exemplary embodiment, the method and system include connecting the L switches of the L processors among the external ports of the L switches in an extended torus architecture and setting the connected L switches thereby interconnecting each of the partitions as a torus.

Method and system of interconnecting processors of a parallel computer to facilitate torus partitioning

A method and system for promoting fault tolerance in a multi-node computing system that provides deadlock-free message routing in the presence of node and/or link faults using only two rounds and, thus, requiring only two virtual channels to ensure deadlock freedom. A lamb set of nodes for use in message routing is introduced, with each node in the lamb set being used only as points along message routes, and not for sending or receiving messages.

System and method for fault tolerance in multi-node system

A method and a system are presented for generating differentially compressed output from binary sources. Given two versions of the same file as input streams, a compact encoding of one of the input streams is generated, by representing it as a set of changes with respect to the other input stream. Algorithms for differencing files requiring time linear in the size of the input and a constant amount of space for execution are presented. In addition, advanced techniques for improving existing differencing algorithms are developed and applied to previous methods. These techniques allow algorithms to increase their efficiency without a loss of compression and to accept arbitrarily large inputs without sacrificing correctness or degrading the compression data rate. The differential compression methods provide a computationally efficient compression technique for applications that generate versioned data.

System and method for differential compression of data from a plurality of binary sources

An information storage system selects target segments for garbage collection only if their age in the information storage system exceeds an age threshold value and, once past the age threshold, in the order of least utilized segments first. The system determines the age of a segment by determining the amount of time a segment has been located in direct access storage devices (DASD) of the information storage system and considers a segment for garbage collection only after the segment has been located in DASD for the selected age threshold value. From the set of candidate segments, the system chooses one or more for garbage collection in the order in which they will yield the most free space. The free space yield is determined by utilization data, so that the least utilized segments are garbage-collected first.

Growing community of inventors

San Jose, CA, United States of America

Larry Joseph Stockmeyer