Difficult Memory Choices In AI Systems


The number of memory choices and architectures is exploding, driven by the rapid evolution in AI and machine learning chips being designed for a wide range of very different end markets and systems. Models for some of these systems can range in size from 10 billion to 100 billion parameters, and they can vary greatly from one chip or application to the next. Neural network training and infer... » read more

ResNet-50 Does Not Predict Inference Throughput For MegaPixel Neural Network Models


Customers are considering applications for AI inference and want to evaluate multiple inference accelerators. As we discussed last month, TOPS do NOT correlate with inference throughput and you should use real neural network models to benchmark accelerators. So is ResNet-50 a good benchmark for evaluating relative performance of inference accelerators? If your application is going to p... » read more

GDDR6 Memory For Life On The Edge


With the torrid growth in data traffic, it is unsurprising that the number of hyperscale data centers has grown apace. According to analysts at the Synergy Research Group, in July of this year there were 541 hyperscale data centers worldwide. That represents a doubling in the number since 2015. Even more striking, there are an additional 176 in the pipeline, so the breakneck growth in hyperscal... » read more

Machine Learning Enabled High-Sigma Verification Of Memory Designs


Emerging applications and the big data explosion have made memory IPs ubiquitous in modern-day electronics. Specifically, the demand for memories with low-die area, low voltage, high capacity, and high performance is rising for use by data center and cloud computing servers. This is essential to serve the exponentially growing connectivity boom and the latest emerging 5G based systems, includin... » read more

Slower Metal Bogs Down SoC Performance


Metal interconnect delays are rising, offsetting some of the gains from faster transistors at each successive process node. Older architectures were born in a time when compute time was the limiter. But with interconnects increasingly viewed as the limiter on advanced nodes, there’s an opportunity to rethink how we build systems-on-chips (SoCs). ”Interconnect delay is a fundamental tr... » read more

Cerfe Labs: Spin-On Memory


Arm has spun off one of its more intriguing semiconductor research projects, a new non-volatile memory type called correlated electron materials RAM (CeRAM) that holds the potential to substantially reduce the cost of memory in everything from edge devices to high-performance computing. Headed by two former Arm Research insiders — Eric Hennenhoefer, who will serve as CEO and Greg Yeric, wh... » read more

Have Processor Counts Stalled?


Survey data suggests that additional microprocessor cores are not being added into SoCs, but you have to dig into the numbers to find out what is really going on. The reasons are complicated. They include everything from software programming models to market shifts and new use cases. So while the survey numbers appear to be flat, market and technology dynamics could have a big impact in resh... » read more

Productivity Keeping Pace With Complexity


Designs have become larger and more complex and yet design time has shortened, but team sizes remain essentially flat. Does this show that productivity is keeping pace with complexity for everyone? The answer appears to be yes, at least for now, for a multitude of reasons. More design and IP reuse is using more and larger IP blocks and subsystems. In addition, the tools are improving, and mo... » read more

DRAM, 3D NAND Face New Challenges


It’s been a topsy-turvy period for the memory market, and it's not over. So far in 2020, demand has been slightly better than expected for the two main memory types — 3D NAND and DRAM. But now there is some uncertainty in the market amid a slowdown, inventory issues and an ongoing trade war. In addition, the 3D NAND market is moving toward a new technology generation, but some are enc... » read more

From Data Center To End Device: AI/ML Inferencing With GDDR6


Created to support 3D gaming on consoles and PCs, GDDR packs performance that makes it an ideal solution for AI/ML inferencing. As inferencing migrates from the heart of the data center to the network edge, and ultimately to a broad range of AI-powered IoT devices, GDDR memory’s combination of high bandwidth, low latency, power efficiency and suitability for high-volume applications will be i... » read more

← Older posts Newer posts →