Securing Accelerator Blades For Datacenter AI/ML Workloads


Data centers handle huge amounts of AI/ML training and inference workloads for their individual customers. Such a vast number of workloads calls for efficient processing, and to handle these workloads we have seen many new solutions emerge in the market. One of these solutions is pluggable accelerator blades, often deployed in massively parallel arrays, that implement the latest state-of-the-ar... » read more

Chip Challenges In The Metaverse


The metaverse is pushing the limits of chip design, despite uncertainty about how much raw horsepower these devices ultimately will require to deliver an immersive blend of augmented, virtual, and mixed reality. The big challenge in developing these systems is the ability to process mixed data types in real time while the data moves uninterrupted at lightning speed. That requires the integra... » read more

Week In Review: Auto, Security, Pervasive Computing


Automotive, mobility The BMW Group will invest $1.7 billion in its U.S. operations to build electric vehicles and batteries, mostly in South Carolina. BMW will drop $1 billion in its South Carolina plant for EV production and $700 million for a new battery-assembly facility in the state. BMW also agreed to purchase battery cells from Japan-based Envision AESC, which plans to construct a new ba... » read more

A Security Maturity Model For Hardware Development


With systems only growing more sophisticated, the potential for new semiconductor vulnerabilities continues to rise. Consumers and hardware partners are counting on organizations meeting their due diligence obligations to ensure security sensitive design assets are secure when products are shipped. This is an iterative process, so a security maturity model is a critical element in getting it ri... » read more

Week In Review: Auto, Security, Pervasive Computing


Automotive, Mobility Hyundai announced all of its vehicles will be software-defined vehicles (SDVs) by 2025. The company said all newly launched Hyundai vehicles will be able to receive over-the-air software updates next year, and that it expects to register 20 million vehicles to its Connected Car Services system by 2025. Hyundai also said it will invest the equivalent of more than $12 billio... » read more

Beyond Autonomous Cars


As the automotive industry takes a more measured approach to self-driving cars and long-haul trucks for safety and security reasons, there is a renewed focus on other types of vehicles utilizing autonomous technology. The list is long and growing. It now includes autonomous trains, helicopters, tractors, ships, submarines, drones, delivery robots, motorcycles, scooters, and bikes, all of whi... » read more

Security For SoC Interfaces Takes Center Stage In Data Protection


Due to today’s connected world, a high volume of valuable data, susceptible to tampering and physical attacks, is processed, stored, and moved between devices, cars, and data centers. And the number of connections continues to grow. Even with supply chain disruptions and the overarching effects of the COVID-19 pandemic on chip manufacturing, the number of global IoT connections grew by 8% in ... » read more

Simplifying AI Edge Deployment


Barrie Mullins, vice president of product at Flex Logix, explains how a programmable accelerator chip can simplify semiconductor design at the edge, where chips need to be high performance as well as low power, yet developing everything from scratch is too expensive and time-consuming. Programmability allows these systems to stay current with changes in algorithms, which can affect everything f... » read more

Put A Data Center In Your Phone!


Datacenters heavily leverage FPGAs for AI acceleration. Why not do the same for low power edge applications with embedded FPGA (eFPGA)? It’s common knowledge for anyone connected to the cloud computing industry that data centers heavily rely on FPGAs for programmable accelerators enabling high performance computing for AI training and inferencing. These heterogeneous computing solution... » read more

AI ASICs Will Become Increasingly Application-Specific


Back in 2017, I blogged about AI ASICs being not exactly ASICs. One of the primary reasons for not calling AI acceleration chips ASIC is because historically ASIC or Application Specific Integrated Circuit has referred to a fixed hardware block with limited programmability. AI ASICs on the other hand offer significant programming via frameworks such as Tensorflow and the point was that they are... » read more

← Older posts Newer posts →