dcn-eu logo
Story image

AWS EC2 strengthens compute power for ML applications

24 Sep 2019

Amazon Web Services now offers Amazon Elastic Compute Cloud (EC2) G4 Instances for organisations that require cloud compute for machine learning and graphics-intensive applications.

Amazon EC2 will leverage the G4 instances to boost resources required to tackle computationally demanding tasks that could benefit from additional GPU acceleration.

The G4 instances feature the latest-generation NVIDIA T4 GPUs, as well as custom 2nd Generation Intel Xeon Scalable (Cascade Lake) processors, up to 100 Gbps of networking throughput, and up to 1.8 TB of local NVMe storage.

According to Amazon Web Services (AWS), G4 instances can provide cost-effective machine learning for applications, such as adding metadata to an image, object detection, speech recognition and more.

Additionally, G4 instances can support graphics-intensive applications such as photorealistic design, video transcoding, and cloud-based game streaming.

AWS explains, “Machine learning involves two processes that require compute – training and inference.”

“Training entails using labelled data to create a model that is capable of making predictions, a compute-intensive task that requires powerful processors and high-speed networking. Inference is the process of using a trained machine learning model to make predictions, which typically requires processing a lot of small compute jobs simultaneously, a task that can be most cost-effectively handled by accelerating computing with energy-efficient NVIDIA GPUs.”

AWS compute services vice president Matt Garman says the company focuses on delivering services that help customers take advantage of compute-intensive applications.

 “AWS offers the most comprehensive portfolio to build, train, and deploy machine learning models powered by Amazon EC2’s broad selection of instance types optimized for different machine learning use cases. With new G4 instances, we’re making it more affordable to put machine learning in the hands of every developer. And with support for the latest video decode protocols, customers running graphics applications on G4 instances get superior graphics performance over G3 instances at the same cost.”

Customers with machine learning workloads can launch G4 instances using Amazon SageMaker or AWS Deep Learning AMIs, which include machine learning frameworks such as TensorFlow, TensorRT, MXNet, PyTorch, Caffe2, CNTK, and Chainer. G4 instances will also support Amazon Elastic Inference in the coming weeks. AWS says this will allow developers to dramatically reduce the cost of inference by up to 75% by provisioning just the right amount of GPU performance.

Customers with graphics and streaming applications can launch G4 instances using Windows, Linux, or AWS Marketplace AMIs from NVIDIA with NVIDIA Quadro Virtual Workstation software preinstalled.

A bare metal version will be available in the coming months. G4 instances are available in the Asia Pacific (Seoul and Tokyo), US East (N. Virginia, Ohio), US West (Oregon, N. California), and Europe (Frankfurt, Ireland, London) Regions, with availability in additional regions planned in the coming months.

G4 instances are available to be purchased as On-Demand, Reserved Instances, or Spot Instances.

Story image
OVHcloud US releases new line of virtual private servers
“Our latest VPS solution is extremely easy to set-up and manage empowering users to fully-leverage the flexibility that accompanies virtualisation."More
Story image
US oil & energy providers hit by plunging market cap in 1H 2020
As the COVID-19 coronavirus pandemic continues to lead many market sectors into turbulence, the energy sector has not escaped unscathed.More
Story image
Fastly adds greater functionality to serverless computing solution
The new features bring customisable logging, real-time and historical metrics, and end to end request visibility to organisations to give developers greater transparency into what happens after code is deployed within a serverless architecture. More
Story image
Google Cloud and HCL strengthen partnership, bolster e-commerce CX
Google Cloud will now be the preferred cloud platform for HCL Commerce, providing a global edge and expanded infrastructure to power businesses’ e-commerce strategies. More
Story image
Colt extends SD-WAN offering for customers moving to cloud
“By using Colt’s SD WAN multi-cloud offering, enterprises can now benefit from better performance, security, scalability and efficiency by accessing all their cloud services through a streamlined infrastructure."More
Story image
Supermicro's SuperServer ready to tackle 'biggest AI, analytics challenges'
“This new SuperServer provides a powerful and trusted foundation for your most secure and mission critical workloads in enterprise, cloud-scale, or hybrid environments.”More