VCore Express Server

The Latest in GPU Computing

The NextIO vCORE™ Express GPU computing solution delivers "must have" features for
the technical and enterprise workgroup computing space including ECC memory for uncompromised accuracy and scalability, and 7X the double precision performance compared to NVIDIA Tesla 10-series GPU computing products. Compared to typical multi-core CPUs, Tesla 20-series computing systems deliver equivalent performance at 1/10th the cost and1/20th the power consumption.

Powerful and Dense Solution

Designed with four Fermi-based NVIDIA Tesla 20-series computing processors with up to 24GB of total GPU RAM in a standard 1U chassis. The vCORE™ Express scales to solve
the world's most important computing challenges - more quickly and accurately.
The vCORE Express achieves:
- Denser Server Clusters - Separating GPUs from servers over high speed PCI Express,
more dense server GPU clusters can be realized in the data center.
- Easier Lifecycle Management - GPUs and servers each have different lifecycles. vCORE Express allows the GPU and the server to be managed separately while minimizing downtime.
- Lower Total Cost of Ownership - Compared to multi-core CPU clusters, vCORE Express
is about a 1/10th of the cost-factoring in the number of servers and infrastructure for equivalent performance to GPU computing systems. When compared to servers with embedded GPUs, the savings associated with lifecycle management will often justify the price of a vCORE system.

GPU Memory Interface 384-bit
GPU Memory Bandwidth Express 2075: 148 GB/sec, Express 2090: 177 GB/sec
Double Precision Floating Point Performance 1.3 Tflops (Peak)
Single Precision Floating Point Performance 2.06 Tflops
Total Dedicated Memory* 24GB GDDR5 / 12GB for M2050
Power Consumption 900W TDP (Typical)
System Interface PCIe x16 Gen2