Posts Tagged ‘high performance computing

  • HP announced this week what it said was deployment of the world’s largest commercial High Performance Computing data center. European aircraft manufacturer Airbus deployed two of the vendor’s 40ft data center containers stuffed with HP-made HPC gear at its sites in Toulouse, France, and Hamburg, Germany.


  • Ed Turkel, business development manager for service providers and HPC at HP, said Airbus deployed the product as a general computing resource for a variety of research and engineering applications. The system ranked 29th on the most recent Top500 Supercomputer Sites list, which came out in June.


  • HPC has not been the most popular use for the HP Pod. So far, the vertical that’s consumed most HP containers has been the service-provider industry. “That’s the place where probably the biggest roll-out [of HP Pods] has been,” Turkel said about the service-provider space.The key reason both HPC customers and service providers have been buying the solution has been the same: speed of deployment. “It becomes a way of building out data center capacity in smaller modular chunks and refreshing those modular chunks as technology changes,” Turkel said.


  • By deploying the two Pods, Airbus doubled its supercomputing power. According to Turkel, it took HP about four months to deliver the solution and connect it to power, water and network from the moment Airbus signed the contract. HP manufactured the aircraft maker’s Pods at a plant in Scotland.HP has both 40ft and 20ft containers. Since the company launched the 20ft solution in February 2010, however, it has not sold as well as the 40ft version has, according to Turkel. HP continues to be tight-lipped about total sales numbers for the product, as do other container vendors.

read more


  • Searchlight Solutions Ltd. today announces its subsidiary, Linux Labs International, Inc., is launching a beta program for IaaS (Infrastructure as a Service) style private and hybrid clouds based on their proprietary Indigo file system and Nimbus cloud operating system. The platform provides a single interface that lets users access high performance computing infrastructure resources available in private clouds.
  • Linux recently completed a key upgrade of its proprietary NimbusOS for use as the Operating System for its High Performance Cloud Computing (“HPC2”) service. Linux Labs NimbusOS was developed to be compatible and configurable for almost all HPC applications. The integration of the proprietary Indigo file system will provide customers a cost effective way to store their data in addition to the high performance processing services offered as part of the Company’s HPC2 service.
  • This beta program is another major step forward towards full deployment of the Company’s complete suite of HPC2 services.

read more

Action items:

  • Develop applications that are elastic in nature and can scale with relative ease.
  • Apps should be low in latency. High security apps, for example, can take a lot of resources and can slow systems down. HPC speeds things up; within the cloud the infrastructure needs to offer sufficient support.
  • The HPC architecture should be based on “loose coupling.” That way, each computer note is independent and has its own storage data. To increase scale, the number of nodes are increased.

July 2018
« Dec    

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 6 other followers


%d bloggers like this: