AWS Announces General Availability of Amazon EC2 DL1 Instances

Today, Amazon Web Services, Inc. (AWS), an Amazon.com, Inc. company (NASDAQ: AMZN), announced general availability of Amazon Elastic Compute Cloud (Amazon EC2) DL1 instances, a new instance type designed for training machine learning models. DL1 instances are powered by Gaudi accelerators from Habana Labs (an Intel company) to provide up to 40% better price performance for training machine learning models than the latest GPU-powered Amazon EC2 instances. With DL1 instances, customers can train their machine learning models faster and more cost effectively for use cases like natural language processing, object detection and classification, fraud detection, recommendation and personalization engines, intelligent document processing, business forecasting, and more. DL1 instances are available on demand via a low-cost pay-as-you-go usage model with no upfront commitments. To get started with DL1 instances, visit aws.amazon.com/ec2/instance-types/dl1.

Machine learning has become mainstream as customers have realized tangible business impact from deploying machine learning models at scale in the cloud. To use machine learning in their business applications, customers start by building and training a model to recognize patterns by learning from sample data, and then apply the model on new data to make predictions. For example, a machine learning model trained on large numbers of contact center transcripts can make predictions to provide real-time personalized assistance to customers through a conversational chatbot. To improve a model's prediction accuracy, data scientists and machine learning engineers are building increasingly larger and more complex models. To maintain prediction accuracy and high quality of the models, these engineers need to tune and retrain their models frequently. This requires a considerable amount of high-performance compute resources, resulting in increased infrastructure costs. These costs can be prohibitive for customers to retrain their models at the frequency they need to maintain high-accuracy predictions, while also posing an obstacle to customers that want to begin experimenting with machine learning.

New DL1 instances use Gaudi accelerators built specifically to accelerate machine learning model training by delivering higher compute efficiency at a lower cost compared to general purpose GPUs. DL1 instances feature up to eight Gaudi accelerators, 256 GB of high-bandwidth memory, 768 GB of system memory, 2nd generation Amazon custom Intel Xeon Scalable (Cascade Lake) processors, 400 Gbps of networking throughput, and up to 4 TB of local NVMe storage. Together, these innovations translate to up to 40% better price performance than the latest GPU-powered Amazon EC2 instances for training common machine learning models. Customers can quickly and easily get started with DL1 instances using the included Habana SynapseAI SDK, which is integrated with leading machine learning frameworks (e.g. TensorFlow and PyTorch), helping customers to seamlessly migrate their existing machine learning models currently running on GPU-based or CPU-based instances onto DL1 instances, with minimal code changes. Developers and data scientists can also start with reference models optimized for Gaudi accelerators available in Habana’s GitHub repository, which includes popular models for diverse applications, including image classification, object detection, natural language processing, and recommendation systems.