You are currently browsing the tag archive for the ‘performance’ tag.

ComputeAvgExtraChart

Jim Connors published a great blog post a little while ago, describing the benefits of using the new Java SE 8 language and API features on multicore embedded systems like the Boundary Devices BD-SL-i.MX6 (built around the Freescale i.MX6 quad-core architecture). 

As you can see in the above diagram, leveraging parallel Java SE 8 Lambda and stream operations can dramatically improve overall performance. What’s more, the effort on the part of the developer is minor and the resulting Java code is portable and scalable across all Java SE 8 platforms, including completely different underlying architectures such as ARM on Linux vs. x86 on Windows.

Compare that to a traditional embedded multicore application, which typically is built with specialized frameworks, knowledge, and tuning and results in platform-specific code with limited portability and scalability. Plus, potentially hard-to-find multithreading bugs …

Out-of-the-box multicore support is another great example of why Java Embedded makes a lot of sense for embedded solutions.

Cheers,

— Terrence

The latest release of Java SE Embedded 6 u21 contains some pretty significant performance enhancements, resulting in up to 3.2x performance over the latest Android 2.2 release on the same hardware.

Check out the detailed benchmarking done by Bob Vandette.

Note that the benchmarks used aren’t even exercising parallel scalability yet. With the multi-core support featured in Java SE Embedded we expect to pull away from Android even further.

Cheers,

— Terrence

Enter your email address to subscribe to this blog

Join 176 other followers

Twitter Updates

February 2020
S M T W T F S
« May    
 1
2345678
9101112131415
16171819202122
23242526272829

Blog Stats

  • 279,717 hits