Amazon Web Services, Inc, an Amazon.com, Inc. company announced the general availability of Amazon Elastic Compute Cloud (Amazon EC2) Trn1 instances powered by AWS-designed Trainium chips. Trn1 instances are purpose built for high-performance training of machine learning models in the cloud while offering up to 50% cost-to-train savings over comparable GPU-based instances.
Also Read: Orange Bank deploys real-time sanctions screening with SAS and Neterium
Trn1 instances provide the fastest time to train popular machine learning models on AWS, enabling customers to reduce training times, rapidly iterate on models to improve accuracy, and increase productivity for workloads like natural language processing, speech and image recognition, semantic search, recommendation engines, fraud detection, and forecasting. There are no minimum commitments or upfront fees to use Trn1 instances, and customers pay only for the amount of compute used.
“We are training large language models that are multi-modal, multilingual, multi-locale, pre-trained on multiple tasks, and span multiple entities (products, queries, brands, reviews, etc.) to improve the customer shopping experience”
More customers are building, training, and deploying machine learning models to power applications that have the potential to reinvent their businesses and customer experiences. These machine learning models are becoming increasingly complex and consume ever-growing amounts of training data to help improve accuracy. As a result, customers must scale their models across thousands of accelerators, which makes them more expensive to train. This directly impacts the ability of research and development teams to experiment and train different models, which limits how quickly customers are able to bring their innovations to market.
AWS already provides the broadest and deepest choice of compute offerings featuring hardware accelerators for machine learning, including Inf1 instances with AWS-designed Inferentia chips, G5 instances, P4d instances, and DL1 instances. But even with the fastest accelerated instances available today, training more complex machine learning models can still be prohibitively expensive and time consuming.
New Trn1 instances powered by AWS Trainium chips offer the best price performance and the fastest machine learning model training on AWS, providing up to 50% lower cost to train deep learning models compared to the latest GPU-based P4d instances. AWS Neuron, the software development kit (SDK) for Trn1 instances, enables customers to get started with minimal code changes and is integrated into popular frameworks for machine learning like PyTorch and TensorFlow.
Trn1 instances feature up to 16 AWS Trainium accelerators that are purpose built for deploying deep learning models. Trn1 instances are the first Amazon EC2 instance to offer up to 800 Gbps of networking bandwidth (lower latency and 2x faster than the latest EC2 GPU-based instances) using the second generation of AWS’s Elastic Fabric Adapter (EFA) network interface to improve scaling efficiency. Trn1 instances also use NeuronLink, a high-speed, intra-instance interconnect, for faster training.