- Integrated Azure's Kubernetes \ Storage \ Networks \ VM pools \ and Docker Registry pieces with CodaLab with the help from Patrick Flickinger.
UPDATE: Ingress and cert-manager are working properly now. We are working to get the challenge upload process to work seamlessly. Seems CodaLab is prepending some links and routes with https instead of http and the normal work flow is interrupted. Patrick is working on a fix. - We did indeed move to a GPU resource pool. I also increased our limits for VMs in East-US and East-US2 to make sure there is enough availability.
- All participant dockers have been run an backed up.
- A new docker submission video (tutorial) has been made and Carolyn is making edits to it.
- Reminded Jayashree to send invoices for July\Aug and Sept (new contract).
Planned for next month |
---|
|
Comments |
---|
Things that went well:
|
Comments |
---|
Things that could be improved:
|
Meeting | Date |
---|---|
Bi-weekly Meeting #1 |
|
Bi-weekly Meeting #2 |
|
Bi-weekly Meeting #e |
Milestones
Description | Date |
---|---|
|
|
Task
Description | Resolution | Status | Creation Date | Close Date |
---|---|---|---|---|
Get sample pneumonia algorithm running on Azure VM | Need to increase quota | complete |
|
|
Get pools of VMs in Azure integration to be based on GPU machines | Depends on above task | pending |
|
|
Risks
Description | Mitigation | Rank | Status | Creation Date | Realization | Close Date |
---|---|---|---|---|---|---|
If we can't get GPU machines hooked up to Kebernetes in Azure then we can't run algorithms | Execute above tasks and get help from Patrick and or Azure support team | -- | closed |
| We needed to request additional access to GPU machines through the Azure service portal |
|
Ingress service in kubernetes is not behaving like the standard tutorials or documentation | Keep plugging away to find the correct configuration | 1 | open |
| Azure has inconsistent docs on how to set this service up. | -- |