# GKE dynamic TPU slicing reaches general availability

> Google enables flexible allocation of reserved TPU resources.

By BIGCHANGE.AI Editorial

Published: 2026-11-10T02:48:38.193Z
Updated: 2026-11-10T02:48:38.193Z
Canonical: https://bigchange.ai/blog/google-gke-dynamic-tpu-slicing-availability

![Cables run from a rack-mounted network panel toward another server cabinet.](https://bigchange.ai/api/media/file/tpu-network-concept.png)
AI-generated illustration for BIG CHANGE

Google Cloud [announced](https://cloud.google.com/blog/products/containers-kubernetes/gke-dynamic-slicing-on-tpus) general availability of GKE dynamic TPU slicing on November 9, allowing jobs to form logical slices from provisioned capacity.

## The big change

- **The change:** Jobs can use different slice shapes.
- **Why it matters:** Teams manage shared resources across workloads.
- **What to watch:** Eligibility depends on the infrastructure.

## How slices are formed

Google says the baseline consists of pre-provisioned 16-VM Ironwood cubes. Super-slicing connects cubes using optical switches; sub-slicing partitions a cube for smaller workloads. Its [documentation](https://docs.cloud.google.com/kubernetes-engine/docs/concepts/dynamic-slicing) gives an example: two `4x4x4` pools form one `4x4x8` slice.

If a component fails, the system can reform a slice using healthy capacity; a usable spare is still needed. 

## The reservation and scheduling constraints

[All Capacity mode](https://docs.cloud.google.com/tpu/docs/all-capacity-overview) gives customers more direct control over reserved TPU resources. In [managed mode](https://docs.cloud.google.com/tpu/docs/capacity-modes), host recovery is automatic. All Capacity exposes the reserved capacity and makes the customer responsible for managing healthy spares. A reservation remains an infrastructure commitment, not extra hardware created for free when a job starts.

Dynamic slicing requires Ironwood TPU7x, an All Capacity reservation and a GKE Standard cluster in the Rapid channel with Container-Optimized OS. Super-slicing requires GKE 1.35.2-gke.1842000 or later; sub-slicing requires 1.36.0-gke.3712000 or later. Google says it is offered in TPU regions, subject to quota and capacity.

Workload admission is a separate concern. [Kueue’s documentation](https://kueue.sigs.k8s.io/docs/concepts/admission/) explains that it checks quota and resource types before reserving resources. Flexible slice shapes do not guarantee immediate admission for every job.

## Sources

- [GKE dynamic slicing](https://cloud.google.com/blog/products/containers-kubernetes/gke-dynamic-slicing-on-tpus) — Google’s announcement.
- [About GKE dynamic slicing](https://docs.cloud.google.com/kubernetes-engine/docs/concepts/dynamic-slicing) — Google documents configurations and prerequisites.
- [All Capacity mode overview](https://docs.cloud.google.com/tpu/docs/all-capacity-overview) — Google explains reservation control and managed-mode differences.
- [Kueue workload admission](https://kueue.sigs.k8s.io/docs/concepts/admission/) — Kueue describes quota checks and resource reservation.