Super Micro Computer is demonstrating supercomputing servers optimized for oil and gas industry applications at the 80th Society of Exploration Geophysicists (SEG) International Exposition 2010 at the Colorado Convention Center, booth 2195 in Hall C, today through October 22.

Supermicro is showcasing several GPU supercomputing servers optimized for the latest NVIDIA M2070 and C2070 Fermi GPU cards with double the onboard memory of the M2050 and C2050 cards. The 1U SuperServer 6016GT-TF-FM207 provides up to 1 TeraFLOPS of double-precision performance, and the SuperServer 7046GT-TRF-FC407 comes loaded with four double-width NVIDIA Fermi C2070 GPU cards and supports three additional PCI-E add-on cards for high-bandwidth I/O. The company is also demonstrating a 2U Twin2 system that supports four hot-pluggable dual-processor (DP) compute nodes, 24 hot-swap SAS2/SATA drives, onboard 40Gb/s InfiniBand, and redundant power.

"As the leader in GPU and Twin supercomputing system architectures, our servers feature exceptional compute density, multiple x16 non-blocking native Gen2 PCI-Express connectivity, and Platinum (94%+ efficiency) Level power supplies to deliver maximum performance-per-watt and provide the best total cost of ownership (TCO)," said Charles Liang, CEO and president of Supermicro. "Equipped with the industry's most efficient thermal designs, our Twin and highly parallel, multi-GPU systems are optimized to increase the speed and accuracy of oil and gas exploration applications including seismic data analysis, reverse time migration, subsurface imaging and data visualization."

Considered among the fastest 1U servers in the world, Supermicro's 6016GT-TF-FM207 Fermi-based server provides the industry's highest compute density and serves as a uniform building block for large-scale deployments. Optimized for performance and reliability, the 6016GT-TF-FM207 supports dual Intel Xeon 5600/5500 series processors and features two NVIDIA Fermi M2070 GPU cards via two Gen2 PCI-Express x16 connections. Also in a 1U form factor, Supermicro's 1022GG-TF server not only supports two twelve-core AMD Opteron 6100 Series processors and two double-width GPU cards, but also can support two 40Gb/s QDR InfiniBand networking cards. In addition, Supermicro's next-generation GPU-optimized supercomputing servers are coming soon.

Suitable for both cluster configurations and personal supercomputing, the 7046GT-TRF-FC407 is a 4U system housed in Supermicro's new rackmount convertible tower chassis, the SC747TQ-R1400. This chassis supports up to 11 full-height, full-length expansion cards, eight hot-swap 3.5" SAS/SATA drives, and special design features that bolster graphics and computationally intensive applications.

The SuperServer 2026TT-H6IBQRF is a 2U Twin2 server featuring an innovative Supermicro architecture in which the four compute nodes, 24 hard drives and two power supplies, are all hot-swappable to facilitate easy maintenance and eliminate downtime, while also saving power and space by sharing the same chassis and power supplies. Each computing node features onboard QDR InfiniBand for 40Gb/s high-bandwidth connectivity and supports up to six hot-swap 2.5" SAS2/SATA drives to deliver unprecedented I/O performance.