Intel® Select Solutions with HPC & AI Converged Clusters
Simplify deployment with a solution that enables HPC batch schedulers to run all workloads—including simulation and modeling, big data analytics, and AI—on a common HPC infrastructure.
The Intel® Select Solutions for HPC & AI Converged Clusters Configuration
Base and Plus configurations for Intel® Select Solutions with HPC & AI Converged Clusters.
| Ingredient | Intel® Select Solutions for HPC & AI Converged Clusters Configuration Details |
|---|---|
| Application node | |
| CPU | 2 x Intel® Xeon® Gold 6248R processor (or higher) |
| Memory | 384 GB DRAM per node |
| 4 x compute nodes | |
| CPU | 2 x Intel® Xeon® Gold 6248R processor (or higher) |
| Memory | 384 GB DRAM |
| Storage (boot) | 240 GB Intel® SSD DC S4610 |
| Storage (capacity) | HPC parallel file system (470 megabits per second [Mbps] per client) |
| Network fabric | Omni-Path Host Fabric Interface Adapter 100 Series |
| Batch scheduler | Open source Magpie on SLURM |
| Software | CentOS Linux installation ISO (minimal or full) 7 build 2003 Intel® Cluster Runtimes 2020.2 Intel® Cluster Checker 2019.9 or higher Intel® HPC Platform RPM packages for EL7 Software version 2018.0 OpenHPC1.3.9 or later Omni-Path Fabric Software (includes Omni-Path HFI Driver) Software version 10.10.3.1.1 or later Intel® OpenVINO™ toolkit software version 2020.4 Intel® Parallel Studio XE Cluster Edition 2020 Apache Spark version 2.4.6 Intel® Optimization for TensorFlow v2.3.0 Horovod v0.19.0 Alluxio v2.3.0 |