Scale node storage based on pod ephemeral-storage requests
**Tell us about your request** What do you want us to build? Enable Karpenter to dynamically scale the size of block device attached to a node at provision time. The size of blocked device would be based on the sum of ephemeral-storage requests of pods being bin-packed to a node plus some overhead. **Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?** Currently, a node's storage capacity can be defined through the Karpenter provisioner through [Block Device Mappings](https://karpenter.sh/v0.16.0/aws/provisioning/#block-device-mappings). This works well, but forces customers to define a static value for all instances launched through a given provisioner. Customers would like the ability to dynamically scale storage of the nodes based on the pod workload or by instance-type. **Are you currently working around this issue?** This can be worked around by defining [Block Device Mappings](https://karpenter.sh/v0.16.0/aws/provisioning/#block-device-mappings) in the Karpenter Provisioner. These values are static for the given provisioner, however, and cannot be dynamically scaled up/down. Related issues: #2512 #2298 #1995 #1467 #3077 #3111 **Additional context** Anything else we should know? **Attachments** If you think you might have additional information that you'd like to include via an attachment, please do - we'll take a look. (Remember to remove any personally-identifiable information.) <!-- Please keep this note for t
Posted by dewjam on Aug 29, 2022, 8:25 PM
Tags
Related issues
- The issue is really the extreme difficulty of programming them, especially for anything sequential (rather than combinatorial).
- I probably could have beared with it if it was only limited to programming, but it showing up in creative fields has made me feel all sor...
- I hate that I'm in deep enough to know what you mean.
- The biggest problem with “AI” is that big corps are shoving it down our throats.
- in the middel of a coding session with 5.6 sol.
- [Hero Run] 535B-A23B on 18T tokens
Raise real-world problems, find people facing the same issue, and discuss solutions.2,500 issues0 comments
reliability (Other)
The issue is really the extreme difficulty of programming them, especially for anything sequential (rather than combinatorial).
I probably could have beared with it if it was only limited to programming, but it showing up in creative fields has made me feel all sor...
I hate that I'm in deep enough to know what you mean.
The biggest problem with “AI” is that big corps are shoving it down our throats.
in the middel of a coding session with 5.6 sol.
[Hero Run] 535B-A23B on 18T tokens
I hate that this is how AI is being used..
inexperienced protobuf user here -- why do non-googlers generally use protobuf?
Scale node storage based on pod ephemeral-storage requests
**Tell us about your request** What do you want us to build? Enable Karpenter to dynamically scale the size of block device attached to a node at provision time. The size of blocked device would be based on the sum of ephemeral-storage requests of pods being bin-packed to a node plus some overhead. **Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?** Currently, a node's storage capacity can be defined through the Karpenter provisioner through [Block Device Mappings](https://karpenter.sh/v0.16.0/aws/provisioning/#block-device-mappings). This works well, but forces customers to define a static value for all instances launched through a given provisioner. Customers would like the ability to dynamically scale storage of the nodes based on the pod workload or by instance-type. **Are you currently working around this issue?** This can be worked around by defining [Block Device Mappings](https://karpenter.sh/v0.16.0/aws/provisioning/#block-device-mappings) in the Karpenter Provisioner. These values are static for the given provisioner, however, and cannot be dynamically scaled up/down. Related issues: #2512 #2298 #1995 #1467 #3077 #3111 **Additional context** Anything else we should know? **Attachments** If you think you might have additional information that you'd like to include via an attachment, please do - we'll take a look. (Remember to remove any personally-identifiable information.) <!-- Please keep this note for t
Comments (0)
No comments yet. Start the discussion.