Friday, March 24, 2006

IBM and grid

from http://www-1.ibm.com/grid/about_grid/ibm_grid/index.shtml

Overview
IBM has a long and thorough involvement with both the technology and the business issues that have led to the grid computing evolution. "Virtualization" — the driving force behind grid computing — has been a key factor since the earliest days of electronic business computing. IBM put the main in mainframe, in part, by creating virtual memory, virtual storage and the virtual processor. This development enabled the computer to do many processing jobs simultaneously for hundreds and eventually thousands of users. Users got mainframe-strength computing; businesses got greater leverage from an expensive and powerful asset.

Fast forward to today. Almost every organization is sitting on top of enormous, unused computing capacity, widely distributed. This is an intolerable situation for customers. (Imagine an airline with 90% of its fleet on the ground, an automaker with 40% of its assembly plants idle, a hotel chain with 95% of its rooms unoccupied.) Once again, virtualization can help.

Grid computing represents this advanced development in virtualization — and IBM Grid Computing continues IBM's history of IT innovation for business. Taking a significant role in the growing grid community, IBM offers a full line of products and services that continue to be developed for both grid customers and those ready for next steps.

Tuesday, March 21, 2006

sybase IQ index

Sybase IQ stores data very differently from conventional RDBMSs. A copy of the data is stored in a column-oriented format (instead of a row-oriented format, as in a conventional RDBMS).
The data values are indexed in arrays of bits for each column, called Bit-Wise indexes. The index either contains the distinct values if there are a few (up to 1,000), or it contains slices of values for larger numbers of distinct values. In both cases, these structures can be manipulated through simple binary operations. These Bit-Wise indexes can be applied to all types of data, for data element

Friday, March 10, 2006

moore bound

An upper limit on the number of nodes in a regular graph of degree d>2 and diameter k:

  N(d,k) <= d(d-1)^k - 2
------------
d-2