Toward Software-Equivalent Accuracy on Transformer-Based Deep Neural Networks With Analog Memory Devices


Abstract:  "Recent advances in deep learning have been driven by ever-increasing model sizes, with networks growing to millions or even billions of parameters. Such enormous models call for fast and energy-efficient hardware accelerators. We study the potential of Analog AI accelerators based on Non-Volatile Memory, in particular Phase Change Memory (PCM), for software-equivalent accurate i... » read more

Power/Performance Bits: Nov. 24


Flexible, low power phase-change memory Engineers at Stanford University created a flexible phase-change memory. The non-volatile phase-change memory device is made up of germanium, antimony, and tellurium (GST) between two metal electrodes. 1s and 0s represent measurements of electrical resistance in the GST material. “A typical phase-change memory device can store two states of resis... » read more

More Data, More Memory-Scaling Problems


Memories of all types are facing pressures as demands grow for greater capacity, lower cost, faster speeds, and lower power to handle the onslaught of new data being generated daily. Whether it's well-established memory types or novel approaches, continued work is required to keep scaling moving forward as our need for memory grows at an accelerating pace. “Data is the new economy of this ... » 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

Memory Issues For AI Edge Chips


Several companies are developing or ramping up AI chips for systems on the network edge, but vendors face a variety of challenges around process nodes and memory choices that can vary greatly from one application to the next. The network edge involves a class of products ranging from cars and drones to security cameras, smart speakers and even enterprise servers. All of these applications in... » read more

Taming Novel NVM Non-Determinism


New memory technologies may have non-deterministic characteristics that add calibration to the test burden — and may require recalibration during their lifetime. Many of these memories are in development as a result of the search for a storage-class memory (SCM) technology that can bridge the gap between larger, slower memories like flash and faster DRAM memory. There are several approache... » read more

DRAM Scaling Challenges Grow


DRAM makers are pushing into the next phase of scaling, but they are facing several challenges as the memory technology approaches its physical limit. DRAM is used for main memory in systems, and today’s most advanced devices are based on roughly 18nm to 15nm processes. The physical limit for DRAM is somewhere around 10nm. There are efforts in R&D to extend the technology, and ultimate... » read more

The Next New Memories


Several next-generation memory types are ramping up after years of R&D, but there are still more new memories in the research pipeline. Today, several next-generation memories, such as MRAM, phase-change memory (PCM) and ReRAM, are shipping to one degree or another. Some of the next new memories are extensions of these technologies. Others are based on entirely new technologies or involve ar... » read more

Cloudy Outlook Seen For IC Biz


After a slowdown in the first half of 2019, chipmakers and equipment vendors face a cloudy outlook for the second half of this year, with a possible recovery in 2020. Generally, the semiconductor industry began to see a slowdown starting in mid- to late-2018, which extended into the first half of 2019. During the first half of this year, memory and non-memory vendors were negatively impacted... » read more

Using Memory Differently To Boost Speed


Boosting memory performance to handle a rising flood of data is driving chipmakers to explore new memory types and different ways of using existing memory, but it also is creating some complex new challenges. For most of the semiconductor design industry, memory has been a non-issue for the past couple of decades. The main concerns were price and size, but memory makers have been more than a... » read more

← Older posts Newer posts →