Modeling AI Inference Performance


The metric in AI Inference that matters to customers is either throughput/$ for their model and/or throughput/watts for their model. One might assume throughput will correlate with TOPS, but you’d be wrong. Examine the table below: The Nvidia Tesla T4 gets 7.4 inferences/TOP, Xavier AGX 15 and InferX 1 34.5. And InferX X1 does it with 1/10th to 1/20th of the DRAM bandwidth of the ... » read more

Revving Up For Edge Computing


The edge is beginning to take shape as a way of limiting the amount of data that needs to be pushed up to the cloud for processing, setting the stage for a massive shift in compute architectures and a race among chipmakers for a stake in a new and highly lucrative market. So far, it's not clear which architectures will win, or how and where data will be partitioned between what needs to be p... » read more

Technological Dead Ends


Sometimes something comes along that looks like it is a portent of things to come, but then turns out to be a technological dead end. For example, in the 1990s, it seemed that you couldn't go to the mailbox or rent a video without getting an AOL CD offering a free trial. They were even in some cereal boxes. It was the era of the 56Kb dialup modem, and AOL's walled garden was king as everyone we... » read more

Emulation Fills The Pre-Silicon Verification Gap For Autonomous Vehicles


Veloce emulators provide the scale and performance to ensure that automotive applications run smoothly, safely, and securely. This paper describes how emulation is used to run realistic driver scenarios, investigate vehicle dynamics, and analyze power and communications metrics — all in a platform that virtualizes the design and allows both hardware and software to be tested together or separ... » read more

Securing The Modern Vehicle


For far too long, we’ve lacked the data needed to fully understand how effective the automotive industry is at addressing the software security risks inherent in connected, software-enabled vehicles. Synopsys and SAE International partnered to commission this independent survey of the current cybersecurity practices in the automotive industry to fill this information gap. Click here to rea... » read more

Managing Analog Designs For Successful Tapeouts


Managing analog designs beyond data management to IP reuse and beyond in order to create a collaborative platform for design management from concept-to-GDSII. Click here to read more. » read more

Simultaneous Localization And Mapping


Amol Borkar, senior product manager at Cadence, explains how to track the movement of an object in a scene and how to match features from one image to the next using SLAM. The technology is used in everything from mobile phones to automotive and drones. » read more

eFPGA as Silicon Debugger


A variety of debugging features are available by implementing the Flex Logix embedded FPGA cores. This includes: • High observability—the many eFPGA IOs allow for thousands of signals to be monitored • Event detection—the eFPGA logic can be configured to detect very complex signal patterns to help identify meaningful events • High visibility—data can be logged by the eLA inside t... » read more

Layered Security Offers Best Embedded Hardware Security


The importance of building a secure and rigid foundation in SoC and system designs has never been so true as it is now, given not only the garden variety of software vulnerabilities existing today, but also micro-architectural attacks on CPUs like Meltdown, Spectre, and Foreshadow. Design engineers should embrace two security-related tenets when selecting a security processor: one is depth o... » read more

Week in Review: IoT, Security, Autos


IoT/Edge Achronix teamed up with Bittware to develop a smart accelerator card based on a 7nm FPGA from Achronix. The card is targeted for edge devices, where pre-processing and acceleration of data movement is critical due to the enormous quantity of data being generated by sensors. The strategy is to move the processing closer to the data, rather than processing input from multiple sensors in... » read more

← Older posts Newer posts →