Last week Penguin Computing announced they had launched two new workgroup-size clusters. First up, a 16 TFLOPS Tesla-based system
Penguin is now offering an Altus 1702 fully-integrated cluster with four twin 1U compute nodes, four NVIDIA Tesla S1070 GPU Computing Systems, Gigabit Ethernet and Penguin’s Scyld management software for $44,985 (configuration Z170034). This system provides over 16 teraflops of compute power in a 9U rack configuration.
Next, and big brother for the little guy that doubles the performance
A cluster option expands the configuration to eight Altus 1702 twin compute nodes and eight Tesla S1070 GPU Computing Systems with over 32 teraflops of performance for less than $89,000. All systems have AMD Opteron 2376 CPU’s and 8GB RAM per node, a 24-port GigE switch and a Scyld annual subscription included for each node. Each Tesla S1070 1U GPU Computing System has four Tesla T10 GPUs, each delivering almost one teraflop of single precision and 80 gigaflops of double precision performance.
More in the release.