February 24, 2010
DHAHRAN, Saudi Arabia, Feb. 24 -- The EXPEC Computer Center (ECC) installed a new supercomputing system in late 2009 with a huge disk storage capacity to support Saudi Aramco's exploration in the frontier areas and the Red Sea. The installed storage of 1,050 terabytes is the largest storage installation in Saudi Aramco's history -- exceeding one petabyte.
"ECC provides the technical computing required by Exploration and Producing (E&P) to maintain a capacity of 12 million barrels of oil per day and to meet the demands for nonassociated gas," said AbdulAziz AbdulKarim, ECC general manager.
"The deployment of this system with the largest storage capacity in the history of Saudi Aramco is another successful example of ECC delivering state-of-the-art technology to optimize oil and gas exploration programs and field development, and to improve recovery."
ECC's storage also delivers high performance. It can move more than 60 gigabytes per second using the latest high-speed, high-density disk technologies. That technology takes up much less space, too. The new 1,050 terabytes of storage fits in 12 racks, compared to 27 racks required for the 650 terabytes of the storage it replaced, while requiring less power and cooling.
"This major installation of new technology and the extensive experience of our professionals helped us to surpass a remarkable milestone on data migration process automation, which was completed four times faster than it would have taken only a year ago," said Exploration Applications Services Department manager Adel Naji. "This new cluster and the vast storage associated with it are essential to support ambitious Saudi Aramco exploration plans."
The supercomputing cluster has more than 200 of the latest processors, each with four processing cores, which together deliver 5.9 teraflops (trillions of floating point operations per second) of performance. Processing jobs on the new cluster are completed in less than half the time it would take on the old cluster.
This cluster is in addition to ECC's lineup of supercomputing systems, three of which are ranked among the world's top 500 supercomputers.
The new cluster introduces an emerging trend in high-performance computing: the graphics processing unit, or GPU. GPUs are commonly used for high-end visualization, but this cluster positions ECC for GPU-based supercomputing.
Evaluation projects are under way to see whether the technology can be used to accelerate Exploration and Producing's computer-intensive applications.
The new cluster includes 72 GPUs, each with 240 cores for processing and delivering one teraflop of single-precision, floating-point performance each -- that is 1,000 times the capability of the old Cray X-MP supercomputer.
"The new serial cluster with expanded disks will have an immediate impact on the Geophysical Data Processing Division's (GDPD) ability to respond to Saudi Aramco's exploration in the Red Sea and the Rub' Al-Khali, as well as to provide accurate seismic images of existing fields for characterization and development," said Exploration Operations Department manager Saleh Al-Maghlouth.
"The new cluster significantly expands GDPD's ability to access data quickly and to apply technologically advanced algorithms to seismic data. This will, in turn, improve the accuracy of subsurface reservoir images and minimize exploration risk. This is important in complex areas like the Red Sea, where imaging beneath the salt bodies will be crucial."
"The installation of this huge system and the migration of hundreds of terabytes of data was a complex and challenging task," said project leader Sadeq Zahir. "It was the excellent collaboration between all project team members that made it a success. It was a privilege to lead this project, which delivered new capabilities to our Exploration users."
Source: Saudi Arabian Oil Co.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?