Compute Engine release notes

Generally available: The storage-optimized Z4D machine series is generally available for Compute Engine. Powered by AMD EPYC Turin processors and Titanium offload processors, Z4D instances are purpose-built for low core usage and high storage density workloads such as:

• SQL, NoSQL, and vector databases
• Data analytics and data warehouses
• Search
• Parallel file systems for AI/ML

The Z4D machine series delivers up to 3 TB of memory and 42,000 GiB of local Titanium SSD capacity. Z4D also supports up to 400 Gbps of network bandwidth using two physical NICs. Z4D instances are available in predefined standardlssd and highlssd machine shapes.

For more information, see Storage-optimized machine family.

Public preview: Network-optimized C4N machine type with 375 GiB to 12,000 GiB of attached Titanium SSD are now available in Preview. You don't have to request allowlist approval.

For more information, see C4N machine series.

Preview: You can create regional disks, including Hyperdisk Balanced High Availability volumes, from custom and public OS images.

For more information, see Create and manage regional disks.

Preview: You can view Workload Manager rule findings directly on the Compute Engine Overview page in the Google Cloud console. The Workload Manager findings tile lets you identify configuration risks for your compute instances and resources against best practices for reliability and security.

For more information, see View Workload Manager rule findings in Compute Engine.

Generally available: You can convert a single-project reservation into a shared reservation, or a shared reservation into a single-project reservation. Modify the share type for a reservation to share your reserved resources with other projects in your Google Cloud organization, or to restrict access to only the reservation's owner project. For more information, see Modify the share type for a reservation.

Generally available: Red Hat Enterprise Linux (RHEL) images pre-configured with Logical Volume Manager (LVM) partitioning are available for Compute Engine in the Google Cloud console. LVM-enabled images let you manage volumes flexibly and size partitions dynamically on your boot disk.

For more information, see Red Hat Enterprise Linux (RHEL).

Generally available: Red Hat Enterprise Linux (RHEL) Extended Update Support (EUS) images are available for Compute Engine in the Google Cloud console. RHEL EUS images let you lock your VMs to a specific minor version while receiving critical security updates and bug fixes for up to two years.

For more information, see Red Hat Enterprise Linux (RHEL).

Generally available: The network- and memory-optimized M4N machine series is generally available. Powered by 5th generation Intel Xeon Scalable processors (Emerald Rapids), M4N instances are purpose-built for network and block storage-intensive workloads such as:

• High-performance vector databases
• Retrieval-augmented generation (RAG) data layers
• Massive in-memory context caching
• Real-time semantic search

The M4N machine series delivers the highest I/O performance available in Compute Engine, supporting up to 400 Gbps of network bandwidth. M4N also offers leading block storage performance with Hyperdisk Extreme that scales up to 25 GiB/s of bandwidth and 1M IOPS. M4N instances are available in predefined machine shapes, ranging in size from 16 to 224 vCPUs and up to 5,952 GB of DDR5 memory.

Preview: Image Builder is available in allowlist-only Preview. Image Builder is a declarative operating system (OS) image customization tool that automates the process of building, customizing, and validating custom OS images within Google Cloud by using Cloud Build. For more information about Image Builder, see About Image Builder.

To onboard your project and request access, fill out the request form.

Preview: Advanced Compute Images provide high-performance images to support your artificial intelligence (AI), machine learning (ML), and high-performance computing (HPC) workloads running on non-GPU instances on Google Cloud.

Advanced Compute Images optimized for CPUs provide a single source of trusted, performance-tuned OS images that remove the need for manual image building for specialized workloads. Each image version is pre-installed with the necessary drivers, network fabrics, and Slurm agents to help you run your workloads.

Generally available: You can use zonal and global extension policies in VM Extension Manager to automatically install and manage extensions, such as the Ops Agent, on a fleet of VMs and to ensure consistent extension states across your project.

To improve observability of enforcement states and guest agent activities, you can view VM extension logs by using Cloud Logging. These logs help you identify and troubleshoot issues with VM extensions.

For more information, see About VM Extension Manager.

Generally available: The U4 Ultra Low Latency (ULL) network-optimized machine family is generally available with allowlist for ULL Solution customers. Designed for financial exchange ecosystems, the U4 machine family provides industry-specific networking capabilities and ultra-low latency unicast and multicast communication. All U4 instances leverage Google's Titanium IPU.

The U4 machine family includes the following machine series:

• U4P machine series: Bare metal instances powered by 5th generation Intel Xeon Scalable processors (Emerald Rapids), delivering specialized compute for exchange operators running core trading systems.
• U4C machine series: Bare metal instances powered by 5th generation Intel Xeon Scalable processors (Emerald Rapids), designed for exchange participants executing latency-sensitive trading strategies.
• U4S machine series: Virtual machine (VM) instances powered by 6th generation Intel Xeon Scalable processors (Granite Rapids), providing advanced flexibility and cost optimization for adjacent workloads and ecosystem services requiring close physical proximity to U4P and U4C instances.

For more information, see the U4 machine family overview in the ULL Solution documentation.

Generally available: Compute flexible committed use discounts (CUDs) are available for G2 and G4 GPU accelerator-optimized machine series. The supported resources include vCPUs, memory, Local SSD disks, and GPUs.

Compute flexible CUDs are spend-based CUDs that apply to eligible Google Cloud spend across Compute Engine, GKE, and Cloud Run. For G2 and G4 machine series, compute flexible commitments provide the flexibility to switch between eligible machine series and regions depending on your workload needs. For GPUs that belong to these machine series, compute flexible commitments don't require attached reservations.

For more information, see Compute flexible CUDs.

A vulnerability (CVE-2026-6726) in the Trusted Computing Group's TPM 2.0 reference implementation code was discovered and is being addressed. For more information, see the GCP-2026-054 security bulletin.

Changed: The minimum provisioned throughput for a Hyperdisk ML volume attached to more than 20 instances is 20 MiB/s per instance, reduced from 100 MiB/s.

For more information, see Share a Hyperdisk ML volume between instances.