HP NVIDIA Tesla K20 5GB GDDR5 Graphics Card C2J97AA

<div><table><tbody><tr><th>Brand</th><td>HP</td></tr><tr><th>Model</th><td>C2J97AA</td></tr></tbody></table></div><strong>Interface</strong><table><tbody><tr><th>Interface<span>&nbsp;</span><i></i></th><td>PCI Express 2.0 x16</td></tr></tbody></table><strong>Chipset</strong><table><tbody><tr><th>Chipset Manufacturer</th><td>NVIDIA</td></tr><tr><th>GPU<span>&nbsp;</span><i></i></th><td>Tesla K20c</td></tr></tbody></table><strong>Memory</strong><table><tbody><tr><th>Memory Size</th><td>5GB</td></tr><tr><th>Memory Type<span>&nbsp;</span><i></i></th><td>GDDR5</td></tr></tbody></table><strong>Details</strong><table><tbody><tr><th>Cooler</th><td>With Fan</td></tr></tbody></table><strong>Features</strong><table><tbody><tr><th>Features</th><td>Reach new heights in performance and system efficiency and open doors for new sets of GPU applications with Kepler GPU architectural enhancements that include the SMX streaming multiprocessor, Hyper-Q technology, and Dynamic Parallelism.<br><br>Maximize performance and reduce data transfers from system memory with large on-board GDDR5 memory, which keeps larger data sets in local memory attached directly to the Kepler GPU.<br><br>Enhance application data integrity with ECC memory protection, which provides critical workstation computing accuracy and reliability by protecting registry files, L1/L2 caches, shared memory, and DRAM.<br><br>Use OpenACC or CUDA programming toolkits for C, C++ or Fortran for parallel application program development.<br><br>Dynamic Parallelism simplifies parallel programming by enabling GPU threads to automatically spawn new threads, eliminating unnecessary program control transfers between the CPU and GPU and allowing the GPU to accelerate a broader set of algorithms.<br><br>Enable multiple CPU cores to simultaneously use the CUDA cores on a single Kepler GPU with Hyper-Q, which boosts average GPU utilization, reduces CPU idle times, and advances programmability.</td></tr></tbody></table>