Tag: cluster compute
While it’s oftentimes worthwhile to present awards to those who are providing a remarkable or innovative service yet still fly under the radar, when it comes to cloud platform leaders, sometimes going with the “obvious” choice is the only option–and for good reason.
Adaptive Computing came forth this morning with some details about their early involvement with Amazon’s new HPC-geared instance type. In addition to the more predictable news about the role of their Moab technology, they also tested Cluster Compute Instances to determine how they would perform for HPC users.
Will average real HPC capability computing applications (e.g. in electronic design automation, automotive applications, or finite-element based material analysis) show a big performance improvement over vanilla EC2 instances, especially those which really demand low latency and high bandwidh?
Following Amazon’s rather unexpected announcement of its new HPC-geared instance type for EC2 called Cluster Compute Instances (CCI) the questions started emerging about the pricing structure. More specifically, about how this news would either answer or complicate the age-old question of buying or renting a cluster — or whether the cloud is the most efficient measure for reducing costs without sacrificing performance.
Amazon’s newly announced HPC offering in the form of Cluster Compute Instances has made big waves in the HPC and cloud space alike this week but as the surprise wears off, some are asking critical questions about the 10GbE decision and the performance in practice.
Reflections on today’s oddly-timed release of Amazon’s Cluster Compute Instances (CCI) which will doubtlessly make the cloud quite a bit more attractive for a wider range of HPC users. It came as quite a suprise to most but is some of the biggest news to unexpectedly hit the wire in quite some time.