Control
In The Trenches With OpenStack
It is hard to believe but it is still early days for cloudy infrastructure. …
Articles by Timothy Prickett Morgan
Compute
HPE Fills A NUMA Server Gap With SGI UV Iron
It takes a lot of engineering talent, time, and money to create custom interconnects that are the basis of scale-up servers, which is one of the reasons that so few companies have been able to sustain that investment over the years. …
Compute
The Inevitable Slide As Server Virtualization Peaks
It has been just about a decade and a half since server virtualization juggernaut VMware debuted the first release of its ESX bare metal hypervisor, and nearly a decade since this hypervisor and its virtual machine guests had enough oomph to take on the job of encapsulating and consolidating multiple operating systems and workloads on since physical servers. …
Compute
Non Volatile Heaps And Object Stores In The Machine
It is great that Hewlett Packard Enterprise is taking the storage bull by its horns with The Machine and working to create a complete compute and storage platform for the day when the needed technologies will all come together. …
Compute
Formula One Racing Can Drive CFD Innovation Faster
Formula One racing teams have the need for speed, and even more importantly, acceleration. …
Compute
Future Systems: What Will Tomorrow’s Server Look Like?
We can talk about storage and networking as much as we want, and about how the gravity of data bends infrastructure to its needs, but the server – or a collection of them loosely or tightly coupled – is still the real center of the datacenter. …
Enterprise
Aiming Object Storage At Hadoop, Hitting Core Enterprise Instead
While compute and networking are fairly homogenous across the enterprise datacenter these days, the same cannot be said for storage, which is a bit of a mess and far less uniform than the clustered storage that is deployed by hyperscalers, cloud builders, and HPC centers. …
Code
How Big Banks Thread The Software Performance Needle
While parallel programming on distributed systems is difficult, making applications scale across multiple machines – or hybrid compute elements that mix CPUs with FPGAs, GPUs, DSPs, or other motors – linked by a network is not the only problem that coders have to deal with. …
Connect
The Long Cycles Of Enterprise Networks
We have been talking about the impending and now delivered 100 Gb/sec speeds for InfiniBand used by HPC centers and for Ethernet to be used by hyperscalers that it is hard to remember sometimes that the enterprise is a significant laggard when it comes to network bandwidth. …
Cloud
How Long Can AWS Keep Climbing Its Steep Growth Curve?
With every new technology that takes off in the IT sector, there is a somewhat predictable curve. …