Update: Lenovo announced plans today to acquire IBM’s x86 server business for $2.3 billion. With the deal, the company would acquire IBM’s System x, BladeCenter and Flex System blade servers and switches, NeXtScale and iDataPlex servers. IBM will keep its System z mainframes, Power Systems, Storage Systems, Power-based Flex servers, and PureApplication and PureData appliances. Note that this […]
“Some of you might be wondering whether Kaveri is good for HPC applications. Compared to discrete GPUs, applications that are already ported and work well on discrete GPUs will continue to be best run on discrete GPUs. However, Kaveri and HSA will enable many more applications to be GPU accelerated.”
Over at Clabby Analytics, Joe Clabby writes that the OpenPOWER Consortium has made progress since it was announced last year. With founding members Google, IBM, Mellanox, and Nvidia, OpenPOWER intends to build advanced server, networking, storage and GPU-acceleration technology aimed at delivering more choice, control and flexibility to developers of next-generation, hyperscale and cloud data centers.
“The last lesson I want to mention is that there is more than one way to do I/O. By switching from the “one-by-one” approach, which wrote the data once the data structure was full, to the “array” method, which created an array of data structures and filled them before writing out the data, I was able to improve my data I/O pattern.”
“In 2014, HPC technology will combine original, un-fragmented data loads with visualization tools to map data—accelerating the creation of relational hypothesis. We will also start to see the first signs of machine learning for enterprise data insights as automated relational analysis enters the Big Data analytics space.”
“At this point, I think it actually really makes sense for the community as a whole to fork Lustre, rename it OpenSFS, remove all the branding IP, and make it GPL/MIT/BSD licensed. Have OpenSFS own the copyrights/IP, have vendors sign up to support it. Sort of like the Linux foundation. And what CentOS does with RedHat.”
“This was a big year for us as we closed with many milestones in what I call a Significant Paradigm shift in Parallelism. The most significant milestone was that the OpenMP Consortium has released OpenMP 4.0 in 2013 with new parallelism features that are productive, portable, and performant across C, C++, and Fortran.”