Share This Post

Cloud News / Google

Google Compute Platform to boost performance with NVIDIA GPUs

Google Compute Platform to boost performance with NVIDIA GPUs

Google Cloud Platform (GCP) recently announced introducing faster Nvidia GPUs (Graphics Processing Units) for Google Compute Engine (GCE).

Firstly, there will be a boost in performance of Google Cloud Platform (GCP) with the public launch of NVIDIA P100 GPUs in beta. Secondly, it has made NVIDIA K80 GPUs generally available on GCE and thirdly, there will be sustained use discounts on both cloud GPUs.

Cloud GPUs provide flexibility, high performance and cost-effectiveness as compared to traditional solutions. The user can customize CPU, memory, disk and GPU configuration as needed.

Cloud GPUs available in passthrough mode provide bare-metal performance. For better disk performance, up to 3 TB of local SSD can be attached to any GPU virtual machine, optionally. It provides per-minute billing and sustained use discounts that reduces the costs.

“For certain tasks, [NVIDIA] GPUs are a cost-effective and high-performance alternative to traditional CPUs. They work great with Shazam’s core music recognition workload, in which we match snippets of user-recorded audio fingerprints against our catalog of over 40 million songs. We do that by taking the audio signatures of each and every song, compiling them into a custom database format and loading them into GPU memory. Whenever a user Shazams a song, our algorithm uses GPUs to search that database until it finds a match. This happens successfully over 20 million times per day.”— Ben Belchak, Head of Site Reliability Engineering, Shazam

GCP had launched GPUs in February this year to benefit customers with extra computation by providing accelerated workloads on machine learning tools. And now GCP has taken another step towards making itself the optimal place for any hardware-accelerated workload.

NVIDIA Tesla P100, based on Pascal GPU architecture, is capable of accelerating the workloads by up to 10 times as compared to K801. It is cost effective and increases the throughput with low number of instances.

Also read: Google offers network flexibility to the end customer with new Network Service Tiers

The Sustained Use Discounts lower the prices of VMs by up to 30% when used to run sustained workloads. The NVIDIA Tesla P100 and K80 GPUs are now available in four regions worldwide.

Image source: Google Cloud Platform blog

The official announcement can be read here.

Rate this post

Share This Post

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Skip to toolbar