Alibaba's MNN Deep Learning Framework Continues Squeezing More Performance

Written by Michael Larabel in Programming on 9 January 2021 at 09:03 AM EST. 1 Comment
PROGRAMMING
Alibaba developers released an updated version of their "blazing fast" lightweight deep learning framework MNN, or the Mobile Neural Network.

This deep learning framework that has been "battle-tested by business-critical use cases in Alibaba" continues working to exploit every bit of possible performance.

With the 1.1.1 release noting more optimizations for "speed up" as well as bug fixes, our MNN test profile for benchmarking with the Phoronix Test Suite and OpenBenchmarking.org was updated against the new version.
MNN Comparison

MNN Comparison

MNN Comparison

MNN Comparison

Indeed, we are seeing some nice speed-ups compared to the state of the code just a few months ago:
MNN Comparison

Inception V3 was the main outlier with mixed results across the systems tested.

More performance numbers can be found via MNN on OpenBenchmarking.org. See how your system performs with Alibaba's deep learning framework by running phoronix-test-suite benchmark mnn.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week