probably means that the particular strand of genes that causes breast cancer was found first by the few scientists. They patent this genetic sequence as their IP. When we are looking for breast cancer abnormality and use a machine that examines gene sequences, it is basically looking for that sth these few cheap ass loserlies patented. If there aint no other gene sequence that causes breast cancer for all of life as we know it, they own practically all remaining hope for the breast cancer afflicted life on earth.
Kunz
Well, much depends on what we use the 100 cores for anyhow. Most of the scientific applications, imaging/video applications run loops, with code sizes in the loops, number of nests etc not very large and in the 100s of lines. Inter/Intra loop dependencies do not occur very often. Basically we have potential for huge available parallelism in code by executing loop iterations in parallel. And we are limited only by the available parallelism in the machine we are using. Also there are compiler code transformations used to transform loop bodies to eek out some of the available parallelism, automatically. Memory system design definitely is a very difficult engineering issue with 100s of cores, best we can do is possibly get cache memory closer i.e. into each core and save on external memory access time etc.
Slashdot Mirror
Rgt. if it is true that they work on marxist-leninits principles i will eat my head.
probably means that the particular strand of genes that causes breast cancer was found first by the few scientists. They patent this genetic sequence as their IP. When we are looking for breast cancer abnormality and use a machine that examines gene sequences, it is basically looking for that sth these few cheap ass loserlies patented. If there aint no other gene sequence that causes breast cancer for all of life as we know it, they own practically all remaining hope for the breast cancer afflicted life on earth. Kunz
Well, much depends on what we use the 100 cores for anyhow. Most of the scientific applications, imaging/video applications run loops, with code sizes in the loops, number of nests etc not very large and in the 100s of lines. Inter/Intra loop dependencies do not occur very often. Basically we have potential for huge available parallelism in code by executing loop iterations in parallel. And we are limited only by the available parallelism in the machine we are using. Also there are compiler code transformations used to transform loop bodies to eek out some of the available parallelism, automatically. Memory system design definitely is a very difficult engineering issue with 100s of cores, best we can do is possibly get cache memory closer i.e. into each core and save on external memory access time etc.
Kunz