This article describes the performance data of AIACC-AGSpeed (AGSpeed) in training models. Compared with the native PyTorch Eager training accelerator, AGSpeed delivers significantly improved performance.
The example in this article tests the performance of AGSpeed when applied to different model training scenarios.
In this example, more than 50 models, including hf_GPT2, hf_Bert, resnet50, and timm_efficientnet, are trained based on the precisions of FP32 and automatic mixed precision (AMP). The following figure shows the performance data of each model in different scenarios.
The following section provides a description of the x-axis and y-axis in the figure.
The following table describes the performance improvement ratio of AGSpeed compared to the native PyTorch Eager. The example in this topic uses throughput as the sole performance indicator. The improvement effect is calculated based on the following formula: Performance improvement ratio = (throughput (AGSpeed) -throughput (Eager)) / throughput (Eager).
Note
The following section shows the performance data of some test models.
Alibaba Cloud Introduces New Pricing Strategy and Service Availability for International Customers
876 posts | 198 followers
FollowAlibaba Cloud Community - April 8, 2024
Alibaba Clouder - June 17, 2020
Alibaba Clouder - June 10, 2020
Alibaba Clouder - July 17, 2020
Alibaba Clouder - September 25, 2020
JJ Lim - April 19, 2023
876 posts | 198 followers
FollowA high-quality personalized recommendation service for your applications.
Learn MoreThis solution provides you with Artificial Intelligence services and allows you to build AI-powered, human-like, conversational, multilingual chatbots over omnichannel to quickly respond to your customers 24/7.
Learn MorePowerful parallel computing capabilities based on GPU technology.
Learn MoreA one-stop service platform for whole-process generative AI engineering and application development, based on Tongyi Qianwen (Qwen) and other popular models
Learn MoreMore Posts by Alibaba Cloud Community