Polski

It is not questionable that Cisco is the biggest networking equipment supplier worldwide. With almost 60% market share, it is clear that a lot of companies is using their hardware. But what if, we want to change question markCisco to something new, or we are just starting Our networking journey, and we are looking for something new and fresh. Is there something as good as Cisco, or even better, when it comes to Wireles
s LAN Networking ?

Continue →

Till today,  most dense HPC systems could accommodate maximum of 4-6 high power GPU NVIDIA accelerators in 2U.

Today, HPE announced new machine called HPE Apollo 6500 server. With two newest Broadwell processors from Intel and eight newest GPU accelerators, it is most dense solution for HPC on the market.

The Apollo 6500 will run NVIDIA GPUs, AMD GPUs, and the Intel Xeon Phi.

It fits well for deep learning, machine leaning and HPC clusters solutions.

HPE Apollo 6500

HPE Apollo 6500, 2U, 8x GP GPU, 2X Intel Xeon E5-2600 v4 Broadwell

 

We will definitly get one of this for testing purposes. If You are interested in pricing, and more details, feel free to contact Us.

Meanwhile You can read more about it, on HPE site

HPE announced new model of Apollo server series. Looking at the architecture, it is clear, that HP didn`t only introduced new chassis which can hold up to two nodes. HPE made big change in the way how disks shelfs are connected to the servers. In Apollo 4510 with oneApollo 4520 node or Apollo 4530 with 3 nodes, every node has access to one of disk shelfs. On the other hand in new Apollo 4520 we get two nodes, which has access to all disks installed in the chassis.

This architecture design is perfect fit for parallel filesystem like Lustre which gives Us very dense solution with HA built-in for OST and MDS servers.

 

 

Detailed look at HP Apollo 4520 specification…

Continue →

We have already written about the capabilities of HP Advanced Power Manager. However, we would like to show an example of practical use of HP APM for monitoring of a cluster, and, more specifically, the power consumption by the cluster. The idea of an application based on APM is a Web page with a cluster map and a diagram of racks.

Continue →

Apparently, users can fill any disk resource. But not in the case of CEPH!
This disk system is built on individual multi-drive servers (nodes) connected to a network. With the addition of subsequent nodes, the system capacity increases linearly and in addition, its efficiency increases. The only limitation is your budget and space in the server room.

Our company has done the production deployment of a very efficient disk cluster using the CEPH system as a shared storage area for the computing cluster. The cluster with a capacity of 1 PB and resistant to failure of any node can reach transfer speeds in excess of 10GBps.

Continue →

One of the most important tasks for every data center and computing cluster administrator is to manage and monitor the existing infrastructure. Often you need a lot of utilities, hardware, and software in order to do this. With HP Advanced Power Manager, you can simplify and hp_apmconsolidate the management and monitoring processes into one device and location.

Continue →