ML and DL Libraries Performance Optimization


ML and DL Libraries Performance Optimization

big data   ·    C++   ·    deep learning   ·    Linux   ·    machine learning   ·    neural networks   ·    Python


A US-based semiconductor company asked Auriga to analyze and optimize Machine Learning and Deep Learning libraries’ performance on new processors.

Projects Highlights

Research and optimization of the ML/DL libraries’ performance: TensorFlow, Caffe, MXNET, scikit-learn, etc.

Low-level analysis of bottlenecks in basic mathematical algorithms (vector/matrix multiplication).

Parallel calculations features.

Comparison with state-of-the-art benchmarks of the leading hardware manufacturers.

Achieved Benefits

Neural networks libraries’ bottlenecks revealed in the course of detailed performance analysis.

Performance maximized due to optimal hardware/software configurations found.

Developed benchmarks demonstrate 20-30% higher performance for deep neural networks training on new processors compared to competing platforms.


Linux  ·  Python  ·  C++  ·  DLBench

DeepBench  ·  Statistical analysis

Most Relevant Cases


MWC Americas 2019: The Era of Intelligent Connectivity

MWC Americas 2019: The Era of Intelligent Connectivity

Mobile World Congress Americas, a large technology conference, was held in Los Angeles, CA in October 2019. Nearly 22,000 attendees from more than 100 countries gathered

Auriga Named a 2019 Global Leader by Clutch

A new study predicts that 20 billion devices will be connected to the IoT by 2020. At Auriga, we offer a wide range of services, including software...

Auriga Attends Intel Experience Day 2019

Intel Experience Day 2019, organized by Intel, one of the major innovative hardware and technology corporations worldwide, took place in Moscow at the end of...