Confirm and Proceed
View More
View Less
System Message
An unknown error has occurred and your request could not be completed. Please contact support.
Reserved - Scan in at least 10 minutes before the beginning of the session or you forfeit your seat.
This has been added to your Planner. Please note this is first come, first served. You have not reserved a seat in this activity.
Waitlisted - You may be assigned a reserved seat if one becomes available.

In order to find repeats of this session, please click on the session title to view the session details.
Personal Calendar
Conference Event
There aren't any available sessions at this time.
System Message
This session is already scheduled at another time. Would you like to...
Please enter a maximum of {0} characters.
{0} remaining of {1} character maximum.
Please enter a maximum of {0} words.
{0} remaining of {1} word maximum.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Replies ()
New Post
Microblog Thread
Post Reply
Your session timed out.
Meeting Summary

ARC302-R - [REPEAT] Patterns for hosting ML models in low latency microservices

Session Description

Deploying deep learning models in a low latency microservices environment is challenging. However, without this, we wouldn't have many of the AI applications that we have today, such as voice response systems, fast image and video analytics, responsive visual search engines, and real-time predictions for business transactions, such as for ride-hailing apps. These challenges are due to the high amount of compute involved, latencies of models and costs, difficulties in deployment, A/B testing, and monitoring performance. In this session, we demonstrate Amazon Elastic Inference, Amazon SageMaker Neo and batch inference, AWS Inferentia, and AWS X-Ray, and we discuss how to use these services to overcome these challenges.

Session Speakers
Additional Information
Chalk Talk
300 - Advanced
Please note that session information is subject to change.
Session Schedule
    Repeat Sessions