This github repo contains examples for deploying common EDA applications with Google Cloud Platform Cluster Toolkit.
- Creating the HPC cluster for EDA workload
- Slurm Cluster - with latest H4D instance(s)
- Storage - Managed Lustre and Filestore
- Managed Lustre
- Filestore
- Synopsys application
- Installation of software
- VCS sample run
- HSPICE sample run
We are using the Google Cloud Platform Cluster Toolkit to create a new cluster.
By using the Cluster Toolkit blueprint yaml:
h4d-slurm-lustre-crd.yaml
This sample blueprint has Slurm setup with Managed Lustre for H4D instances. Chrome Remote Desktop node is also added to this blueprint.
Once the setup of the Cluster Toolkit, one can run the following
./gcluster deploy -w hpc-slurm-h4d.yaml
After the cluster setup (about 45mins), login to the slurm login node and we can see the sinfo and lustre file system:
Slurm cluster:
$ sinfo
PARTITION AVAIL TIMELIMIT NODES STATE NODELIST
h4d* up infinite 2 down# slurmh4d-h4dnodeset-[0-1]
Storage:
$ df -h
Filesystem Size Used Avail Use% Mounted on
devtmpfs 7.7G 0 7.7G 0% /dev
tmpfs 7.7G 0 7.7G 0% /dev/shm
tmpfs 7.7G 8.5M 7.7G 1% /run
tmpfs 7.7G 0 7.7G 0% /sys/fs/cgroup
/dev/sda2 50G 20G 31G 39% /
/dev/sda1 200M 5.9M 194M 3% /boot/efi
tmpfs 1.6G 0 1.6G 0% /run/user/0
10.x.x.x@tcp:/lustrefs 35T 22M 35T 1% /data
10.x.x.x:/homeshare 2.5T 0 2.4T 0% /home
slurmh4d-controller:/opt/apps 50G 20G 31G 39% /opt/apps
tmpfs 1.6G 0 1.6G 0% /run/user/1911945234