Hello everyone,
When running VFVS for 10K and 100K compounds on c5.24xlarge X 10 nodes using qvina02, CPUUtilization of each c5.24xlarge node stayed at 100%. On the other hand, when the number of compounds is increased to 1M or 10M, it stays at around 15%. Setting RootVolume as follows only improved it to 27%.
ComputeSettings:
LocalStorage:
RootVolume:
Size: 1000
Iops: 16000
Throughput: 1000
If tempdir_default=/dev/shm, the job for more than 1M compounds failed due to insufficient storage space, so /tmp is used in this case. When the tempdir_default was changed to /shared/tmp, the utilization dropped to about 5%.
How can we set up various configuration files such as all.ctrl to run at 100% CPUUtilization even when the compound is over 1M?
all.ctrl (13.2 KB)
CPUUtilization is Percentage of allocated EC2 compute units that are currently in use on the instance.
Key Metrics for EC2 Monitoring | Datadog (datadoghq.com)
List the available CloudWatch metrics for your instances - Amazon Elastic Compute Cloud