Skip to content
#605: AWS Trainium-powered Amazon EC2 Trn1n instances episode artwork

Jul 20, 2023 20 min

#605: AWS Trainium-powered Amazon EC2 Trn1n instances

from AWS Podcast Hosted by Amazon Web Services

How to get the best price performance in Amazon EC2 for the most demanding machine learning training workloads? Tune in to learn how AWS Trainium-based Amazon EC2 Trn1n instances can help you train your network-intensive generative AI models at scale. Amazon EC2 Trn1n instances double the bandwidth offered by Trn1 instances to 1600 Gbps of EFA and deliver up to 20% faster time-to-train than Trn1 instances. Both Trn1 and Trn1n instances deliver up to 50% savings on training costs over comparable Amazon EC2 instances. Tune in to learn more about this new launch that helps you increase performance, reduce costs, and also improve energy efficiency when training your large-scale ML models. Trn1 Website: https://go.aws/44v90ST AWS Neuron Website: https://bit.ly/46SLyjX AWS Trainium Website: https://bit.ly/3DqiCSM AWS Inferentia Website: https://go.aws/44ymLA9

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

#605: AWS Trainium-powered Amazon EC2 Trn1n instances

0:00 20:58

No transcript yet.

Requests help us decide what to transcribe next.

Frequently asked questions

How long is this episode of AWS Podcast?

This episode is 20 minutes long.

When was this AWS Podcast episode published?

This episode was published on July 20, 2023.

Can I download this AWS Podcast episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.