Loading...

XML

Word

Printable

Type: Feature
Resolution: Unresolved
Priority: Critical
Fix Version/s: None
Affects Version/s: None
Component/s: Node
Labels:

Activity Type:
Product / Portfolio Work
Parent Link:
OCPSTRAT-1692AI Workloads for OpenShift
Blocked:
False
Blocked Reason:

Hide

None

Show
None
Ready:
False
Size:
None

Target Version:
None
Release Blocker:
None

SFDC Cases Counter:
SFDC Cases Open:
SFDC Cases Links:

PX Review Complete:
None
PX Priority Data:
None
PX Impact Score:
PX Technical Impact:
None
PX Impact Range:
None
PX Scheduling Request:
None
PX Technical Impact Notes:
None

Intelligence Requested:
Market:

Overview
OpenShift's Dynamic Resource Allocation (DRA) for AI Workloads is designed to optimize the management of specialized resources like GPUs, critical for running AI/ML and large language model (LLM) inference workloads. By using flexible, dynamic requests for high-performance hardware, OpenShift empowers users to meet demanding AI application needs with cost-effective, real-time resource scaling and efficient workload scheduling.

incorporates

OCPNODE-2510 DRA Upstream Work

Closed

OCPSTRAT-1756 DRA: Attribute-Based GPU Allocation in OpenShift with NVIDIA GPU operator -TP 4.20

Release Pending

OCPSTRAT-408 Deprecated : Structured parameter in DRA : refer to https://issues.redhat.com/browse/OCPSTRAT-1756

Closed

Assignee:: Gaurav Singh

Reporter:: Gaurav Singh

Need Info From:: None

Contributors:: None

Architect:: None

QA Contact:: None

Doc Contact:: None

Product Operations Engineering Contact:: None

Votes:: 0 Vote for this issue

Watchers:: 6 Start watching this issue

Created:: 2024/11/08 8:52 PM

Updated:: 2025/10/15 5:41 PM

Details

Description

Attachments

Issue Links

Easy Agile Planning Poker

Activity

People

Dates