This Is Google Cloud with Intel

Intel has a 16 year strategic partnership with Google to create joint solutions for digital transformation and business continuity with the best operational efficiencies. Everything needed to build, scale, and accelerate transformation for today and tomorrow’s leading workloads. Intel instances in Google Cloud offer balanced performance to suit virtually every workload: HPC, Big Compute, Analytics, Databases, VMWare, and AI.

Featured Topics

The Solution Your Enterprise Needs to Accelerate Digital Transformation

Achieve performance on your terms: Whether your applications require multi-cloud management, large memory, machine learning, high-performance compute, or HiFi content rendering and delivery, Google Cloud instances with the latest Intel technology provide scalable, mission-critical solutions for each.

Intel and Google

Working Together

Be cloud agile. Intel architecture delivers performance required to respond to evolving business needs. Harness seamless application portability, data mobility, and compelling total cost of ownership benefits across the broadest range of workloads and services – whether compute, networking, or storage.

Intel Technology and Products Support a Full Breadth of Google Cloud Workloads

C2 instances: Near real-time performance for the most demanding workloads

M2 instances: Large SAP certified VMs up to 12TB

Live Migration compatibility between Intel Xeon-based VMs - having this capability is in many instances - global - regions

Deep Learning VMs and containers optimized and configured to support common AI frameworks like TensorFlow.

N2, C2, M2 instances: up to 40% higher performance 2x floating point performance5

Preemptible VMs at 80% price discount6

Availability of free software optimization tools such as Intel Studio XE to reduce time to deployment.

N2, C2, M2 instances: Built-in AI inference performance5## with DL boost instructions.

Operational Efficiency with customizable memory and compute levels

Intel-powered Instance Offerings

Customers can easily build upon Intel-powered Google Cloud instances enabling your data-driven platforms to store more, process faster, and migrate quicker—with a flexible, high-performance set of instances that work with your workloads and your budget, not against it.

Product Table

Instance Type

C2

Compute Optimized

N2

General Purpose

M2-UltraMem and mega-mem

Memory Optimized

N1

General Purpose

M1-UltraMem

Memory Optimized

M1-MegaMem

Memory Optimized

Google Cloud

VMware Engine

Target Workload

Apps Needing Single Threaded Frequency

(i.e. HPC,EDA, Gaming)

Mainstream

High Memory Database (i.e. SAP HANA)

Mainstream

High Memory Database (i.e. SAP HANA)

High Memory Database (i.e. SAP HANA)

VMware-based Instances

CPU

2nd Gen Intel Xeon Scalable

2nd Gen Intel Xeon Scalable

2nd Gen Intel Xeon Scalable

Various Intel Xeon, including Xeon Scalable

Intel Xeon E7v4

Intel Xeon Scalable

2nd Gen Intel Xeon Scalable

vCPUs

Up to 60

Up to 80

208 / 416

Up to 96

40 / 80 / 160

96

72

Memory

Up to 240GB

Up to 624GB

6TB / 12TB

Up to 624GB

1TB / 2TB / 4TB

1.4TB

1TB

All-Core Turbo Boost speed

3.8 GHz (Sustained)

3.4 GHz

3.4 GHz

2.7 GHz (Xeon Scalable)

2.6 GHz

2.7 GHz

3.9 GHz

Baseline Cloud speed

3.1GHz

2.8 GHz

2.5 GHz

2.0 GHz (Xeon Scalable)

2.2 GHz

2.0 GHz

2.6 GHz

Intel AVX-512

(2x floating pt)

Yes

Yes

Yes

Xeon Scalable (using MinCPU platform)

No

Yes

Yes

Intel Deep Learning Boost

Yes

Yes

Yes

No

No

No

Yes

Available in GCP Regions

12

13

5

24

17

10

2

Can take-in AVX-512 and VNNI instructions Yes Yes   Yes      

Intel Select Solution for Google Cloud’s Anthos

Intel Select Solutions for Google Cloud’s Anthos* can simplify your move to a hybrid or multi-cloud solution by offering pre-tested, validated solutions that make use of your existing VMware vSphere* and VMware vSAN* environment. The solution allows you to quickly and easily deploy and manage apps and workloads using virtual machines (VMs) or Kubernetes* containers across your on-premises environment, on your hybrid cloud, or to your preferred cloud provider.

Case Studies

Google Cloud and Intel AI Partner with NASA-FDL On Space Challenges

For 2019 Google Cloud and Intel AI partnered with NASA’s Frontier Development Labs for their fourth annual summer research accelerator. Intel mentors worked with google cloud mentors and early career Ph.D’s from around the world, applying the latest machine learning and AI technology for smarter, faster problem solving on some of space travel’s most pressing issues like lunar resources, space weather and earth observations.

Learn more

Additional Resources

Intel-Google Cloud Alliance:
Customer Benefits

Google NEXT OnAir 2020

Stay Connected


Keep tabs on all the latest news with our monthly newsletter.

产品和性能信息

1The new Compute-Optimized VMs offer a greater than 40% performance improvement compared to current GCP VMs.” Bart Sano, VP of Platforms, Google Cloud. Learn more.
2From Google Cloud’s CPU platforms page: https://cloud.google.com/compute/docs/cpu-platforms: C2 instance all-core turbo CPU frequency is 3.8 GHz, with the lowest first generation N1 machine type base frequency being 2.0 GHz. Note, all-core turbo not available for first generation N1 instances. Improvement from 2.0GHz to 3.8GHz = 90%.
3“The new general-purpose machine types (N2) offer greater than 20% price-performance improvement...compared with first generation N1 machines.” Hanan Yousef, Product Manager of Compute Engine, Google Cloud. Learn more.
4“The new general-purpose machine types (N2) offer greater than 20% price-performance improvement for many workloads and support up to 25% more memory per vCPU compared with first generation N1 machines.” - Hanan Yousef, Product Manager, GCE. Learn more.
5“With double the FLOPS per clock cycle compared to previous-generation Intel® Advanced Vector Extensions 2 (Intel® AVX2), Intel® AVX-512, built into 2nd Gen Intel Xeon Scalable Processors boosts performance and throughput for the most demanding computational tasks in applications, such as modeling and simulation, data analytics and machine learning, data compression, visualization, and digital content creation.” Learn more.
6“Preemptible VMs are up to 80% cheaper than regular instances.” Google Cloud Blog, Preemptible Virtual Machineshttps://cloud.google.com/preemptible-vms/