feat(infra): Update Vespa to optimized configuration

[justin-tahara]

Description

More memory optimized instances for Vespa Cloud deployment. Optimizing our usage of instances and increasing bin-packing.

Referencing the configurations in here: https://cloud.vespa.ai/en/reference/aws-flavors.html

<resources vcpu="8.0" memory="128Gb" architecture="arm64" storage-type="local" disk="475Gb"/>

[Provide a brief description of the changes in this PR]
Instance upgrade for higher utilization

How Has This Been Tested?

[Describe the tests you ran to verify your changes]
QA through Vespa workflow

Backporting (check the box to trigger backport action)

Note: You have to check that the action passes, otherwise resolve the conflicts manually and tag the patches.

• This PR should be backported (make sure to check that the backport attempt succeeds)
• [Optional] Override Linear Check

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
internal-search	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Jul 28, 2025 5:49pm

[greptile-apps]

Greptile Summary

This PR optimizes the Vespa Cloud deployment configuration by reducing the node count from 75 to 60 while increasing memory per node from 64GB to 128GB and slightly bumping disk storage from 474GB to 475GB. The changes are made specifically to the cloud-services.xml.jinja template file which defines the resource allocation for Vespa's cloud deployment.

The modification targets better bin-packing efficiency by using fewer but more powerful memory-optimized instances, following AWS flavor configurations available in Vespa Cloud. This approach actually increases total memory capacity from 4.8TB (75 × 64GB) to 7.68TB (60 × 128GB) - a 60% increase - while potentially reducing operational costs through better resource utilization. The change only affects cloud deployments and leaves local/development configurations unchanged.

This optimization aligns with modern cloud deployment practices of using larger, more efficient instances rather than many smaller ones, which can reduce network overhead, improve cache efficiency, and simplify cluster management.

Confidence score: 4/5

• This is a straightforward infrastructure optimization with clear benefits and minimal risk
• The configuration change is well-documented and references official Vespa Cloud documentation
• Only affects cloud deployment configuration, leaving other environments untouched

_{1 file reviewed, no comments}

_{Edit Code Review Bot Settings | Greptile}