ReleaseBytes
gcp Google Cloud Blog ·

Google Cloud's M4N VM family is now generally available for I/O and memory-bound workloads

blogdatagcpgaarchitecthealthcaremediagcp-compute-engine
feature

Google Cloud has announced the general availability of its M4N virtual machine family, designed for I/O-intensive, high-memory workloads. These VMs offer industry-leading per-core IOPS and throughput, addressing bottlenecks for applications like Oracle databases, SAP HANA, and large-scale analytics. The M4N family can reduce Oracle database TCO by over 20% by enabling right-sizing of compute resources and improving cache efficiency. Powered by 5th Gen Intel Xeon Scalable processors and Google Cloud's Titanium architecture, M4N instances deliver up to 25 GiB/s host storage and 400 Gbps VM-to-VM network bandwidth.

  • →Google Cloud M4N VM family now generally available
  • →High performance for mission-critical enterprise applications
  • →Reduce TCO and Oracle licensing overhead by over 20%
  • →Superior VM-to-VM and internet egress network bandwidth
  • →Doubled block storage performance with Hyperdisk Extreme
Features (2) ›
  • Google Cloud M4N VM family now generally available

    The M4N series in Google Compute Engine, purpose-built for I/O intensive, high-memory workloads, is now generally available. It offers the highest per-core IOPS and throughput for high-memory instances among leading hyperscalers, complementing existing memory-optimized families.

  • High performance for mission-critical enterprise applications

    M4N targets workloads demanding extreme RAM and I/O performance, including Oracle, SAP HANA, SQL Server, EHR systems, and generative AI RAG data layers. It offers high memory-to-core ratios (up to 26.57 GB/vCPU) paired with 25 GiB/s storage for rapid data ingestion and transaction logging.

Enhancements (3) ›
  • Reduce TCO and Oracle licensing overhead by over 20%

    M4N's industry-leading high memory-to-vCPU ratio allows users to right-size infrastructure, avoiding over-provisioning of compute cores just for RAM. This can lead to over 20% TCO reduction for Oracle databases compared to similar offerings, by improving cache-hit ratios and optimizing licensing costs.

  • Superior VM-to-VM and internet egress network bandwidth

    Built on Google Cloud's custom Titanium offload architecture, M4N delivers up to 400 Gbps aggregate VM-to-VM network bandwidth and up to 200 Gbps internet egress bandwidth. This enables non-blocking data exchange for distributed database clusters and real-time streaming data layers.

  • Doubled block storage performance with Hyperdisk Extreme

    Paired with Hyperdisk Extreme, M4N instances deliver up to 25 GiB/s aggregate block storage throughput and 1,000,000 IOPS, doubling the performance of current M4 instances. This is beneficial for rapid database recovery, transactional checkpointing, and instant in-memory index reloads.

Read the original announcement →

https://cloud.google.com/blog/products/compute/compute-engine-m4n-vms/

Related releases