Google Agent Compute Engine gives each Gemini agent its own microVM
Original title: Google's Agent Compute Engine gives every Gemini agent its own hardware-isolated micro virtual machine
A preview is planned in the coming months; Google says 200,000 agents suspended and resumed in under 500 milliseconds.
Google announced Agent Compute Engine, which places each AI agent in its own isolated virtual machine. The service already powers Gemini and is set to open for preview in the coming months. Google says each microVM is secured by its Titanium chip, idle VMs pause automatically, and shared memory lowers cost. The same approach underpins open-sourced Agent Substrate, a Kubernetes runtime that Google says can fit ten times as many agents on one server. Oversight additions include token tracking, model routing, and third-party identity providers, extending to Cloud Run, GKE, and managed endpoints.
Unconfirmed: Density, latency, and isolation claims are Google's own statements.
- Idle agent VMs pause and free the machine for other agents.
- Google says Agent Substrate fits ten times as many agents on one server.
- A 200,000-agent test suspended and resumed in less than 500 milliseconds.
- Identity support named by Google includes Ping Identity and SailPoint.
- Vendors
- Organizations
- Ping IdentitySailPoint
- AI models
- Gemini
From the original article
1 min read at the-decoder.com
With Agent Compute Engine, Google is introducing a runtime environment designed specifically for AI agents, while also expanding the governance capabilities that enable companies to manage agents in cloud, hybrid, and on-premises environments. The article Google's Agent Compute Engine gives every Gemini agent its own hardware-isolated micro virtual machine appeared first on The Decoder.
By Matthias Bastian · the-decoder.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.