Sweden’s Royal Institute of Technology to Upgrade XT to XE6
Cray announced details today surrounding their latest order from Sweden’s Royal Institute of Technology (KTH: Kungliga Tekniska Hoegskolan). Earlier this year, KTH purchased a new Cray XT6 supercomputer. This procurement will upgrade the brand new machine to an XE6 platform. The upgrade will bump performance to around 300TF and swap SeaStar networking gear for Cray’s new Gemini network.
We are very pleased that after only a few months, PDC has made the decision to upgrade its Cray XT6m system to our new Cray XE6 supercomputer,” said Dr. Ulla Thiel, Cray vice president, Europe. “Easy upgradeability is an important design element of Cray supercomputers, and this is a great example of a customer leveraging its HPC investment and
…
ALICE at U of Leicester Billed as Green Machine
The University of Leicester has deployed a new supercomputing system, deemed ALICE, that is being billed as a performance and green upgrade from previous systems. The new system and facility cost an estimated £2.2 million in order to make it an efficient operational environment.
Mary Visser, Director of IT services at the university, said: “It’s fascinating to see how researchers work these days — looking for patterns in huge datasets and simulating complex phenomena. Usually, you need to be a real techie to engage with this kind of work. But, we have social scientists and economists with big problems to solve who didn’t sign up to be computer programmers. Our team aims to help
…
University of Minnesota Goes Purple with UV 1000
SGI announced today that the University of Minnesota Supercomputing Institute for Advanced Computational Research (MSI) will soon deploy a new HPC system based around the Altix UV 1000. The new system will be made possible by a grant by the National Institutes of Health [NIH] awarded to the University of Minnesota in June of this year.
From the release:
MSI named the system Koronis after a Minnesota lake; it will feature a 1,152-core SGI Altix UV 1000 system with a shared-memory architecture in which each core can access all 3.1 terabytes of memory directly. Koronis
…
TACC Lands Another Super
The Texas Advanced Computing Center and the University of Texas announced details on the latest in computing gear to enter their Austin campus. The $9 million machine, named “Lonestar”, will land on the J.J. Pickle Research Campus early next year. Wait, don’t they already have a “Lonestar” machine? Indeed they do, but this version is a big upgrade.
The new system, like its predecessors, will be dedicated to open science research. As such, it will join its older [and larger] brother “Ranger” as being served out to Teragrid consumers. How much new research? 302 Tflops worth. Not too shabby.
I think the most interesting thing about this system is that it’s really optimized for achieving real performance on scientific applications,
…
SGI Delivers to Stanford Solar Variability Project
SGI announced today that the Hansen Experimental Physics Laboratory Solar Observatories Group at Stanford University has lined up a full range of SGI gear for their Solar Dynamic Observatory (SDO) group’s research in solar variability. The SDO strives to understand the characteristics of the Sun’s interior and the components of its magnetic activity to help forecast space weather, and relies on SGI’s technology to receive, process and archive massive amounts of data.
For the first time, scientists are able to comprehensively view the dynamic nature of storms on the Sun. The immense amount of data collected during our research requires a powerful HPC solution capable of ingesting and analyzing data quickly and with precision,” said Phil
…
SDSC fields 100 TFLOPS Appro super
SDSC announced late last week that they have been awarded $2.8M to deploy a new 10,368-core super from Appro. SDSC has focused in recent years on systems that enable data intensive computing, with a particular emphasis on flash memory systems. In November last year SDSC announced that Appro would be providing its next $20M super (named Gordon) as a follow on to SDSC’s (apparently successful) Dash flash memory prototype.
“Trestles is appropriately named because it will serve as a bridge between SDSC’s unique, data-intensive resources available to TeraGrid users now and into the future,” said Mike Norman, SDSC’s interim director. “The overarching goal of Trestles is to enable as much productive science as
…
Update on Hawking’s UV
Last week we covered news that the research consortium headed by Stephen Hawking was upgrading to an SGI UV 1000 system. That story was thin on actual details about the deployed system, so I got in touch with SGI to see if we could find out more. Here is what I heard back from a spokesperson
- The system is just shy of 1,000 cores
- Installation is scaling to three racks
- 2+ TB of Memory
- As the consortium members (7 universities) require more computing power, the system will scale with their needs.
- The system was delivered to COSMOS August 18
Not a huge amount of shared memory, given that the system can handle up to 16TB, but it is a lot relative to what it …
Hawking orders a UV 1000
eWeek Europe is reporting today that the UK Computational Cosmology Consortium (COSMOS) in Cambridge is one of SGI’s latest UV 1000 customers.
Stephen Hawking, who heads up that consortium, had this to say
“Recent progress towards a complete understanding of the universe has been impressive, but many puzzles remain,” said Hawking. “Cosmology is now a precise science, and we need supercomputers to calculate what our theories of the early universe predict and test them against observations of the present universe.”
Remember the UV 1000 is the big version of SGI’s Xeon-based shared memory machine. The UV will replace COSMOS’ current Altix 4700. According to SGI there is more to the buy than just hardware
Altix
…
More computing in a big metal box
After years of no one wanting (or willing) to talk about their trailer sales, HP now has two releases pretty close together. Maybe this means that there is something here after all. Everyone with a container offering has always said to me that the sales cycles are much longer than systems sales cycles, because a container is more like a datacenter than a system. Maybe that wasn’t just marketing hoohah.
Following close on the heels of the iVEC deployment in Australia, Purdue has announced that it, too, has signed up to put part of its computing resources in a trailer
Known for its world-leading research in nanotechnology, structural biology and atmospheric chemistry, Purdue is committed in its strategic plan
…
Virginia Tech Lands NSF Grant
Virginia Tech announced today that is was awarded a $2million NSF grant for a new computational platform. The new machine, called HokieSpeed, will feature traditional CPUs and GPUs. Wu Feng, associate professor of computer science as well as electrical and computer engineering and principal investigator on the grant, along with Khidir Hilu, professor of biological sciences, and Scott King, professor of geosciences, all at Virginia Tech, successfully applied for the $2 million grant through the National Science Foundation’s Major Instrumentation Program for the new gear.
HokieSpeed is expected to catalyze new approaches for conducting research via the synergistic amalgamation of heterogeneous supercomputing and cyber-enabled tools to enhance ease of use. In particular, it
…
