HPC Phase I Platform Usage
📄️ Platform Introduction
The first phase of the HPC AI high-performance computing platform was officially put into use in April 2022. The total FP64 computing power of the first phase HPC is 0.246 Pflops (quadrillion calculations per second). The CPU adopts dual Intel Xeon Gold 6348 with 28 cores and a base frequency of 2.6GHz, totaling 16 compute nodes. The GPU uses NVIDIA A30, with a total of 32 GPU cards. The cluster's high-performance network utilizes Mellanox HDR 100G InfiniBand, with high-speed Ethernet using 10Gb/s, and the parallel file system BeeGFS has a storage capacity of 701GB.
📄️ Connecting to the Cluster
Web Portal Login for Phase One Cluster
📄️ Submitting Jobs
Job Template Introduction
📄️ Viewing Job Status via Web Interface
Accessing My Jobs
📄️ Job Management
View Nodes
📄️ Dependent Jobs
Dependent jobs: When a job needs to wait for other jobs to run or finish before starting, you can configure its dependencies. name indicates the dependent job name, and status indicates the status of the dependent job.