The container craze on Linux platforms just took an interesting twist now that Red Hat is shelling out $250 million to acquire its upstart rival in Linux and containers, CoreOS. …
While no one has yet created an exploit to take advantage of the Spectre and Meltdown speculative execution vulnerabilities that were exposed by Google six months ago and that were revealed in early January, it is only a matter of time. …
There is a kind of dichotomy in the datacenter. The upstart hyperconverged storage makers will tell you that the server-storage half-bloods that they have created are inspired by the storage at Google or Facebook or Amazon Web Services, but this is not, strictly speaking, true. …
An unexpected jump in enterprise spending coupled with the ongoing heavy spending by hyperscalers, cloud builders, and communications companies revamping their networks of gear coupled with the ramp of the “Skylake” Xeon SP processors launched last July gave Intel the best overall quarter in its history, gauged by revenues and profits, and the best one also that its Data Center Group has ever posted. …
To get straight to the point: nobody wants to have large grain snapshots of data for any dataset that is actually comprised of a continuous stream of data points. …
Energy is not free, not even to energy companies, and so they are just as concerned with being efficient with their supercomputers as the most penny pinching hyperscaler or cloud builder where the computing is the product. …
If Mellanox Technologies had not begun investing in the Ethernet switching as it came out of the Great Recession, it would be a much different company than it is today. …
If it were not for the insatiable bandwidth needs of the twenty major hyperscalers and cloud builders, it is safe to say that the innovation necessary to get Ethernet switching and routing up to 200 Gb/sec or 400 Gb/sec might not have been done at the fast pace that the industry as been able to pull off. …
Designing and manufacturing processors – or paying a third party foundry to do some of the work – and then manufacturing systems and updating and modernizing operating systems and middleware is tough work. …
When it comes to supercomputing, more is almost always better. More data and more compute – and more bandwidth to link the two – almost always result in a better set of models, whether they are descriptive or predictive. …
All Content Copyright The Next Platform