[Roadmap] Q1 2025

[peterschmidt85]

This issue outlines the major items planned for Q4 2024.

Hardware

• major [Feature]: Support ARM CPUs #2101

On-prem clusters

• major [Feature] Support GPU blocks #1780
• major GPU health checks
• major Large clusters Test and document max number of active runs and instances supported by the server #2183
• medium [Feature] Support head nodes with SSH fleets #2010
• medium Detect firmware info

Cloud clusters

• major AWS Lustre
• major AWS EFA (incl. ensure it works out of the box) [Feature]: Enable all available network cards on AWS instances #1804
• minor Support multi-host TPUs #1337

General features

• major GPU quotas per user/project
• major Support multi-attach volumes #1467
• medium [Feature]: Allow to configure a service healthcheck #2181
• medium Stats metrics in UI
• medium Interactive offer selection (prototype)
• medium [Feature]: Support websockets for services without a gateway #2231
• minor [Feature]: Zero-downtime in-place update for services #2180

UX

• medium [UI] New UI for fleets and instances #2133
• medium In-place update of fleets (cloud and on-prem)
• medium Show deleted resources (incl. runs, jobs, instances, volumes, etc) Show deleted runs in the UI #2150
• minor [Feature]: Allow listing available key resources such as gpu, region, and backends #2142
• minor [UX] Add a short CLI flag to configure retry + on: no-capacity #2169
• minor [Feature]: Show Docker image pull progress #2187
• minor [AMD] Support default Docker image for AMD #2193

Experimental

• minor [Feature]: on-prem gateway #1402

Troubleshooting

• medium [Bug]: dstack marks instance as terminated without terminating it #1551
• minor [Bug]: Git authentication errors aren't propagated to the CLI #2033
• minor [Feature]: Use a dedicated status when exceeding max_duration #1701

Examples

• medium Ataching volumes to user home directories
• medium Using AWS EFA
• medium Auto-scaling services via API