NVIDIA® and Google Cloud Platform

NVIDIA® Tesla® P100 GPUs now available on Google Cloud Platform in Beta

Contact Us

Google Cloud Platform Now Offering NVIDIA® Tesla® P100 & NVIDIA® Tesla® K80 GPUs

Predicting our climate’s future. A new drug to treat cancer. Some of the world’s most important challenges need to be solved today, but require tremendous amounts of computing to become reality.

Together, NVIDIA® and Google Cloud are helping you achieve faster results to address these challenges--all without massive capital expenditures or complexity of managing the underlying infrastructure. With NVIDIA® Tesla® K80 (generally available) and NVIDIA® Tesla® P100 GPUs (beta) now available on Google Cloud Platform, deep learning, analytics, physical simulation, and molecular modeling take hours instead of days.

Benefits of using NVIDIA® Tesla® P100 & K80 GPUs on Google Cloud Platform

Increased Performance for Complex Computing
Increase the speed of your complex, compute-intensive workloads such as machine learning training and inference, medical analysis, seismic exploration, video transcoding and scientific simulations. Easily provision Google Compute Engine instances with NVIDIA® Tesla® P100 or NVIDIA® Tesla® K80 to handle your most complex compute-intensive workloads.
Reduce Costs with Per-Second Billing
With Google Cloud Platform’s per-second pricing, you pay only for what you need, with up to a 30% monthly discount applied automatically. Save on up-front capital expenditure while enjoying the 24/7 uptime and scalable performance you have come to expect from NVIDIA® Tesla® GPUs.
Optimize Workloads with Custom Machine Configurations
Optimize your workloads by precisely configuring an instance with the ratio of processors, memory and NVIDIA® GPUs you need instead of modifying your workload to fit within limited system configurations. Save costs by replacing hundreds of non-accelerated nodes with a single, powerful compute instance with up to eight NVIDIA® GPUs for HPC and AI workloads.
Integrate Seamlessly with Cloud Machine Learning
Tackle the explosion of data generated every day by transactional records, sensor logs, images, videos and more. With NVIDIA® GPU-accelerated cloud computing resources, you can generate insights from your data without the need to move it out of the cloud. NVIDIA® Tesla® P100 and NVIDIA® Tesla® K80 GPUs are tightly integrated with Cloud Machine Learning Engine, dramatically slashing the time to train machine learning models on large datasets using TensorFlow framework, and tightly integrating with Dataflow, BigQuery, Cloud Storage and Datalab.
Accelerate ML Training Time & Deliver Efficient ML Inference
Solving today’s complex challenges with ML requires training exponentially more complex deep learning models in a practical amount of time. NVIDIA® Tesla® P100 GPUs, on Google Cloud Platform dramatically reduce training time for these models from weeks to a few hours, and also provide greater efficiency when running those trained models for inference by providing an order of magnitude higher throughput with low latency for better user experiences.
Build on Google’s Global Infrastructure
Access some of the same hardware that Google uses to develop high performance deep learning products, without having to worry about the capital expenditures or IT operations of managing your own infrastructure. NVIDIA® Tesla® P100 and NVIDIA® Tesla® K80 GPUs on Google Cloud Platform means the hardware is passed through directly to the virtual machine to provide bare metal performance.

NVIDIA® GPUs available on Google Cloud Platform


Unified Supercomputing

Designed to boost throughput and save money for both HPC and ML applications. Powered by the NVIDIA® Pascal™ architecture, each Tesla® P100 delivers 4.7 and 9.3 TeraFLOPS of double-precision and single-precision performance for HPC and ML workloads.

Greater Efficiency with CoWoS with HBM2

Applications often spend more time and energy waiting for data than to process it. The NVIDIA® Tesla® P100 tightly integrates compute and data on the same package by adding Chip on Wafer on Substrate (CoWoS) with HBM2 technology to deliver unprecedented computational efficiency. This integration provides a huge generational leap in application performance by delivering up to 3X memory bandwidth over prior-generation solutions.

Simplified Parallel Programming with the Page Migration Engine

Parallel programming just got a lot simpler with the Pascal architecture. The Page Migration Engine frees developers to focus more on tuning for computing performance and less on managing data movement. It also allows applications to scale beyond the physical memory size of the GPU, with support for virtual memory paging. With Unified Memory technology, developers see a single memory space for the entire instance to dramatically improve productivity.

Try NVIDIA® P100 ON Compute Engine

NVIDIA® Tesla® K80 is generally available in the Google Cloud. It drastically lowers model training times and HPC cost by delivering superior performance with fewer, more powerful server instances, engineered to deliver above 5-10x performance boost on real-world applications.

Over 450 industry-leading HPC applications already support NVIDIA® GPUs, including all top 10 HPC applications and all deep learning frameworks. With features like dual-GPU design and Dynamic GPU Boost, Tesla® K80 is built to deliver superior performance for these applications.

TRY NVIDIA® K80 ON Compute Engine

Additional Resources