Compiling And Optimizing Neural Nets


Edge inference engines often run a slimmed-down real-time engine that interprets a neural-network model, invoking kernels as it goes. But higher performance can be achieved by pre-compiling the model and running it directly, with no interpretation — as long as the use case permits it. At compile time, optimizations are possible that wouldn’t be available if interpreting. By quantizing au... » read more

Week In Review: Design, Low Power


Tools & IP Arm unveiled the Cortex-R82, a 64-bit, Linux-capable Cortex-R processor targeted for next-generation enterprise and computational storage solutions. The Cortex-R82 provides 2x performance depending on workload compared to previous Cortex-R generations and provides access of up to 1TB of DRAM for advanced data processing in storage applications. It offers an optional memory manag... » read more

Apples, Oranges & The Optimal AI Inference Accelerator


There are a wide range of AI inference accelerators available and a wide range of applications for them. No AI inference accelerator will be optimal for every application. For example, a data center class accelerator almost certainly will be too big, burn too much power, and cost too much for most edge applications. And an accelerator optimal for key word recognition won’t have the capabil... » read more

Interconnects Emerge As Key Concern For Performance


Interconnects are becoming increasingly challenging to design, implement and test as the amount of data skyrockets and the ability to move that data through denser arrays of compute elements and memories becomes more difficult. The idea of an interconnect is rather simple, but ask two people what constitutes an interconnect and you're likely to get very different answers. Interconnects are e... » read more

All-in-One Vs. Point Tools For Security


Security remains an urgent concern for builders of any system that might tempt attackers, but designers find themselves faced with a bewildering array of security options. Some of those are point solutions for specific pieces of the security puzzle. Others bill themselves as all-in-one, where the whole puzzle filled in. Which approach is best depends on the resources you have available and y... » read more

The Challenge Of Keeping AI Systems Current


Semiconductor Engineering sat down to discuss AI and its move to the edge with Steven Woo, vice president of enterprise solutions technology and distinguished inventor at Rambus; Kris Ardis, executive director at Maxim Integrated; Steve Roddy, vice president of Arm's Products Learning Group; and Vinay Mehta, inference technical marketing manager at Flex Logix. What follows are excerpts of that ... » read more

Customization And Limitations At The Edge


Semiconductor Engineering sat down to discuss the edge constraints and the need for security with Jeff DeAngelis, managing director of the Industrial and Healthcare Business Unit at Maxim Integrated; Norman Chang, chief technologist at Ansys; Andrew Grant, senior director of artificial intelligence at Imagination Technologies; Thomas Ensergueix, senior director of the automotive and IoT line of... » read more

Performance Metrics For Convolutional Neural Network Accelerators


Across the industry, there are few benchmarks that customers and potential end users can employ to evaluate an inference acceleration solution end-to-end. Early on in this space, the performance of an accelerator was measured as a single number: TOPs. However, the limitations of using a single number has been covered in detail in the past by previous blogs. Nevertheless, if the method of cal... » read more

Big Changes In AI Design


Semiconductor Engineering sat down to discuss AI and its move to the edge with Steven Woo, vice president of enterprise solutions technology and distinguished inventor at Rambus; Kris Ardis, executive director at Maxim Integrated; Steve Roddy, vice president of Arm's Products Learning Group; and Vinay Mehta, inference technical marketing manager at Flex Logix. What follows are excerpts of that ... » read more

Week In Review: Design, Low Power


Perforce Software acquired Methodics. Founded in 2006 and based in San Francisco, Methodics' IP lifecycle management and traceability software will join Perforce's larger portfolio of DevOps software that includes version control, Agile planning, and static code analysis. The two companies have had a strategic partnership in place with customers using software from both companies. Terms of the ... » read more

← Older posts Newer posts →