Cruise Automation is a primary developer of autonomous auto technology. In this session, we will dive into the infrastructure which permits us to run hundreds of thousands of autonomous simulations each and every working day and assess the results swiftly and efficiently. Cruise operates the huge greater part of our testing on Google Cloud, having gain of large scalability of compute and GPU assets for our assorted workloads.
Our simulation frameworks permit us to replay facts gathered from highway tests or generate sophisticated versions on human-described eventualities. We will acquire a search at the technological know-how and infrastructure behind these frameworks, difficulties scheduling and scaling our workload on GCP, complexities accessing substantial-definition 3D maps and other means necessary to operate each individual simulation, and our information pipeline which will allow engineers to rapidly instrument and examine their final results.
Construct with Google Cloud → https://little bit.ly/2KbTers
View much more:
Up coming ’19 Compute Classes below → https://little bit.ly/Upcoming19Compute
Subsequent ‘19 All Classes playlist → https://bit.ly/Upcoming19AllSessions
Subscribe to the GCP Channel → https://bit.ly/GCloudPlatform
Speaker(s): Adrian Macneil
Session ID: CMP207
solution: Cloud – Normal fullname: Adrian Macneil function: Google Cloud Next 2019
