-
Hello Alexander Murdoch,
Good day!
Thank you for your feedback. The question is not wrong. Let’s break down your concerns:
-
“It doesn’t say where HPC is running. I’d assume on-prem.”
— Although the question doesn’t explicitly state where the HPC workloads are running. Given that it mentions an on-premises NAS system supporting these workloads, it’s a reasonable assumption that the HPC workloads are also running on-premises. - “They want to move storage and HPC to the cloud (they mention system). Is this correct?”
— In this context, “System” primarily refers to the NAS (Network-Attached Storage) storage system. The question is focused on moving the storage system to the cloud. While HPC (High-Performance Computing) refers to the workloads or applications that use this storage system. The HPC workloads are not being directly moved but need to be supported by the new cloud-based storage solution. - “They want to maintain low-latency in the company network or in the cloud? This is definitely not clear.”
— The company is looking to move its storage system (NAS) to the cloud. They need to ensure that this cloud-based storage solution can still support their HPC workloads with the same low-latency performance they currently have on-premises. Additionally, low latency is crucial for HPC workloads as it ensures quick data access and processing. The cloud migration aims to leverage the scalability and flexibility of cloud services while not compromising on the speed and efficiency of their current system. -
“Answers mentions EC2, I’d assume those are the ones used once the whole system is migrated.”
— EC2 (Elastic Compute Cloud) instances are mentioned in the answer options because they are the primary compute service in AWS. When migrating an on-premises workload to the cloud, EC2 instances typically replace the physical servers used in the on-premises environment.
In this scenario, the EC2 instances would be the compute resources running the HPC workloads, while the file system (like FSx for NetApp ONTAP) would provide the storage. The placement of these EC2 instances (e.g., in a cluster placement group) can affect the overall performance and latency of the solution.
Therefore, while the question doesn’t explicitly mention EC2, its inclusion in the answer options is a logical extension of moving an on-premises workload to the AWS cloud.
I hope this clarifies your question.
Thank you.
Regards,
Neil @ Tutorials Dojo
-
