Release Notes for EGS Version 1.17.0

Release Date: 13th Feb 2026

The Elastic GPU Service (EGS) platform is an innovative solution designed to optimize GPU utilization and efficiency for your AI projects. EGS leverages the power of Kubernetes to deliver optimized GPU resource management, GPU provisioning, and GPU fault identification.

We continue to add new features and enhancements to EGS.

These release notes describe the new changes and enhancements in this version.

info

• Across our documentation, we refer to the workspace as the slice workspace. The two terms are used interchangeably.
• The EGS Controller is also referred to as the KubeSlice Controller in some diagrams and in the YAML files.
• The EGS Admin Portal is also referred to as the KubeSlice Manager (UI) in some diagrams and in the YAML files.

What's New 🔈

North-South Gateway Visualization

We have introduced North-South (NS) Gateway status visualization in the EGS Portal. The NS Gateway is an essential component that facilitates communication between the slice workspace and the EGS Controller. This enhancement offers improved visibility into the health of the NS Gateway, allowing you to easily monitor its status and identify any potential issues that may arise. This improvement helps ensure seamless communication between the slice workspace and the EGS Controller.

For more information, see:

• View NS Gateway Status from the Admin Portal.
• View NS Gateway Status from the User Portal.

Time Slicing for GPU Provisioning Request

We have introduced GPU Provisioning Request (GPR) Time Slicing, a feature that allows you to specify the duration for which a GPU provision request (GPR) is active. With GPR time slicing, GPU oversubscription is achieved by time-slicing a physical GPU across multiple workloads, ensuring efficient resource sharing, higher utilization, and improved scalability.

For more information, see:

• Manage GPU Requests from the Admin Portal.
• Manage GPU Requests from the User Portal.

Known Issues

Issue: Workload Redistribution with Enable EGS option may not work as intended.

Description: When the Workload Redistribution feature is used with a Workload Placement that has the Enable EGS option enabled, workload redistribution may not operate as intended. The behavior is currently under investigation and will be resolved in a future release.

Impact: Workloads may not be redistributed correctly across worker clusters.

Workaround: Do not enable the Enable EGS option for Workload Placements until this issue is resolved.