SIMD Code Generation - Coder Summit 2018
1,184 views · Published 10 April 2018 · 3:42 · Indexed 31 October 2026
Channel: MATLAB · 2018 · Science & Technology
Embedded Coder® generates native SIMD instructions starting in R2018a. This support offers 10X or more code efficiency improvements by leveraging the parallelism afforded by embedded hardware. Learn more about embedded system solutions: https://goo.gl/kCqw3F Embedded Coder native SIMD code generation is a major improvement upon previous SIMD code generation which used code wrappers. The native SIMD code is faster and of higher qualtiy, and appears like professional handwritten SIMD code. Learn more about Embedded Coder: https://goo.gl/4uLCr2 In some examples, featuring Intel AVX and SSE for Windows and Linux, the code executes 16x faster with SIMD instructions. This presentation is from Coder Summit Talks, which are live recordings of developers and engineers debuting their best MATLAB and Simulink coder optimizations and examples at an annual technical interchange. © 2018 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names maybe trademarks or registered trademarks of their respective holders.
More from this channel
-
1:47
AUTOSAR for Production with Model-Based Design
-
4:31
Simulating in Real Time: Backhoe
-
5:37
Optimizing Shift Schedule to Maximize Fuel Economy
-
3:34
Importing Measured Data to Estimate Fuel Economy
-
5:16
Modeling a Ratchet Mechanism with Leadscrew
-
3:21
Running Parallel Simulations: Aileron Parameter Sweep
-
4:30
Finding Causes of Slow Simulations
-
4:58
Optimizing a Flight Control System