-
Notifications
You must be signed in to change notification settings - Fork 97
[TDX] Added Intel TDX support to docsum helm charts #818
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
1d5cdaa
to
e7f14db
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The implementation lacks changes for docsum-ui. Also tgi and llm-uservice are already enabled.
Please fix the following:
- fix description accordingly to what new component is enabled,
- correct the formatting of the PR description to match the suggested default format,
- fix the sign-off to have only one correct entry with valid email address.
helm-charts/docsum/README.md
Outdated
# helm install docsum docsum --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --values docsum/gaudi-tgi-values.yaml ... | ||
# To use Gaudi device with vLLM | ||
# helm install docsum docsum --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --values docsum/gaudi-vllm-values.yaml .. | ||
# To run chatqna with Intel TDX feature |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
docsum
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
fixed
helm-charts/docsum/README.md
Outdated
# To use Gaudi device with vLLM | ||
# helm install docsum docsum --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --values docsum/gaudi-vllm-values.yaml .. | ||
# To run chatqna with Intel TDX feature | ||
# helm install docsum docsum --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --set tgi.LLM_MODEL_ID=${MODELNAME} --set docsum.tdxEnabled=true --set docsum.resources.limits.memory=1Gi --set docsum-ui.tdxEnabled=true --set docsum-ui.resources.limits.memory=2Gi --set llm-uservice.tdxEnabled=true --set llm-uservice.resources.limits.memory=4Gi --set tgi.tdxEnabled=true --set tgi.resources.limits.memory=80Gi --set whisper.tdxEnabled=true --set whisper.resources.limits.memory=13Gi |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
docsum-ui not enabled
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It is with this line:
--set docsum-ui.tdxEnabled=true --set docsum-ui.resources.limits.memory=2Gi
…n docsum README.md Signed-off-by: Krzysztof Sandowicz <[email protected]>
for more information, see https://pre-commit.ci Signed-off-by: Krzysztof Sandowicz <[email protected]>
Signed-off-by: Krzysztof Sandowicz <[email protected]>
989860d
to
232981f
Compare
Signed-off-by: Krzysztof Sandowicz <[email protected]>
Description
This PR is an extension to #799 with Intel TDX support enabled for all docsum microservices.
Subcharts enabled to be run with Intel TDX:
Issues
n/a
Type of change
Dependencies
n/a
Tests
Tested the deployment of docsum example on the running cluster without and with Intel TDX for: