In hands-on ResNet-50 image-recognition testing, the HPE ProLiant DL380 Gen11 server handled significantly more samples per second than the generation model, while also delivering lower latency.
PT found that on ResNet-50 image-recognition tests using FP32 precision, the HPE ProLiant DL380 Gen11 delivered 2.86 times the inference performance while reducing latency by 30.1 percent compared to the previous model. According to the report, PT “also tested the HPE ProLiant DL380 Gen11 server at lower precision levels, which place greater demand on CPU resources, and found its performance to be strong with both Int8 and bfloat16 precision levels.” These performance results suggest the Gen11 server “could do more work with a given number of servers or could achieve a fixed amount of inference work using fewer servers, which has the potential to lead to savings.”
To read the report, visit https://facts.pt/Jj5UV9r. To see the infographic, visit https://facts.pt/uF1b16G.
About
Principled
Principled
Contact Information:
Principled
828-455-0312
Contact via Email
https://www.principledtechnologies.com
Read the full story here: https://www.pr.com/press-release/904015
Press Release Distributed by PR.com
Copyright © 2024 PR.com and its licensors
, source