Home > Workload Solutions > High Performance Computing > White Papers > Dell Validated Design for Government HPC, Artificial Intelligence, and Data Analytics: AI Inferencing Option > Compute servers
Compute servers are the primary computational resource in an HPC cluster. These servers are responsible for running the preferred workloads. Optimal compute configuration depends on the applications and customer specific goals. There are several configurations available including heterogenous options when environments have mixed workloads. In a traditional HPC environment, many traditional modelling and simulation applications remain CPU only, but they may share resources with newer codes that can take advantage of GPUs for some selective calculations. These data centers may even share resources with separate departments that run entirely GPU dependent workloads, such as deep learning. Such an organization may need to decide between choosing nodes with both CPU+GPU compute capabilities, or creating a smaller subset of servers specifically catered toward GPU performance.
The table below the available compute platforms. The options vary in density and compute type (GPU support). These include AMD-based and Intel-based options, the selection of which depends on customer needs and workload.
Platforms | Dell PowerEdge R6625 |
Dell PowerEdge R660 | |
Dell PowerEdge C6620 | |
Dell PowerEdge XE9680 + H100 GPU | |
Networking | NVIDIA® ConnectX®-6 HDR100 InfiniBand™ adapter |
NVIDIA® ConnectX®-6 HDR InfiniBand adapter | |
NVIDIA® ConnectX®-7 NDR200 InfiniBand™ adapter | |
NVIDIA® ConnectX®-7 NDR InfiniBand adapter | |
Dell PowerSwitch N3248TE (100 GbE) |