What next for symmetric multiprocessing on SoC?
By Peter Clarke, Courtesy of EE Times Europe
Mar 5 2007 (15:26 PM)
LONDON — Despite moves by Intel Corp. and Advanced Micro Devices Inc. to bring their dual and four-core processors into embedded applications, there is a sense that some of the companies that pioneered symmetric multiprocessing in the system-on-chip (SoC) world are finding it harder to make progress.
Some symmetric multiprocessors have been designed to be application specific " PicoChip Ltd. (Bristol, England) Alphamosaic Ltd. (Cambridge, England) are two examples. This helps make software problems tractable. For other SMP architectures, intended to be more general, there is still a generic problem of how to write software for parallel processing.
At present, the trend is for multiple processors or SoCs, each with their own proven software payload, to be brought together to create heterogeneous multiprocessor systems. This combines the advantages of integration and reduced die area with minimum development rework. It allows developers to concentrate on additional features often in the form of algorithms set to run on an additional processor in the heterogeneous SoC. However, there may come a point when systems could benefit from the reduced physical complexity of symmetrical multiprocessors.
To read the full article, click here
Related Semiconductor IP
- NPU IP Core for Mobile
- NPU IP Core for Edge
- Specialized Video Processing NPU IP
- HYPERBUS™ Memory Controller
- AV1 Video Encoder IP
Related News
- Gaisler Research announces Linux 2.6 Symmetric Multiprocessing (SMP) support for the LEON3 processor
- Mentor Graphics Announces Multicore Solutions for Symmetric and Asymmetric Multiprocessing
- MIPS Technologies Announces Symmetric Multiprocessing (SMP) Support for Android Platform on MIPS-Based SoCs
- ARM improves chip's graphics, multiprocessing abilities
Latest News
- Jim Keller: ‘Whatever Nvidia Does, We’ll Do The Opposite’
- FlexGen Streamlines NoC Design as AI Demands Grow
- IntoPIX Presents Its New Titanium Software Suite: Empowering AV-Over-IP Workflows With Speed, Quality & Interoperability
- Global Semiconductor Sales Increase 2.5% Month-to-Month in April
- Speedata Raises $44M to Launch First-Ever Chip Designed Specifically for Accelerating Big Data Analytics - Compute's Second Largest Workload