Optimizing HPC workloads for cost-efficiency with cloud bursting - Considering cloud call

[fonadain]

Hi everyone,

I'm working on a research project that simulates complex weather patterns. My current HPC cluster is handling the load okay, but for peak seasons, the simulations take significantly longer. I'm interested in exploring cloud bursting to handle these spikes.

However, I also need to consider the cost implications. One concern is that the frequent data transfer between my on-premise cluster and the cloud during these bursts could be expensive, especially with large weather datasets.

Here's my question:

• For those using cloud bursting for HPC workloads, how do you manage data transfer costs? Are there specific strategies or cloud provider features that can help minimize these costs, particularly for scenarios involving large datasets?

• As an aside, I'm also curious if anyone has experience using cloud bursting for HPC while integrating with a cloud-based call center solution. While not directly related to data transfer, any insights would be appreciated!

Thanks!

[support_s]

System recommended content:

1. Optimizing facility operation in high density data center environments

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.