HBM2e Offers Solid Path For AI Accelerators


Today, AI processors are so blazingly fast that they’re constantly having to wait for data from memory. Unfortunately, with the status quo, memory is just not fast enough to unleash the true performance of those new and highly advancing AI processors. In simple terms, AI processor performance is rapidly growing, and memory is not keeping up. This creates a bottleneck, or what Rambus calls the... » read more

Resetting Serial Memory When A System Failure Occurs


Stability is an important consideration in embedded design, but electronic systems can suffer malfunctions triggered by a myriad of root causes. These include poor signal integrity, power spikes, software errors, and erroneous user interaction, just to name a few. These anomalies, in turn, can lead to inconsistent system operation or can even cause the system to hang. Although the list of po... » read more

HBM2 Vs. GDDR6: Tradeoffs In DRAM


Semiconductor Engineering sat down to talk about new DRAM options and considerations with Frank Ferro, senior director of product management at Rambus; Marc Greenberg, group director for product marketing at Cadence; Graham Allan, senior product marketing manager for DDR PHYs at Synopsys; and Tien Shiah, senior manager for memory marketing at Samsung Electronics. What follows are excerpts of th... » read more

Accurate Error Bit Mode Analysis Of STT-MRAM Chip With A Novel Current Measurement Module


Authors: (Advantest) Ryo Tamura, Ibuki Mori Naoyoshi Watanabe; (Tohoku University) Hiroki Koike, Tetsuo Endoh. A novel memory test system is needed for future STT-MRAM mass production that supports error bit analysis and its mode categorization on STT-MRAM chip measurement, as STTMRAM cell’s switching is a probabilistic phenomenon based on quantum mechanics. In order to meet this requireme... » read more

Controlling Variability And Cost At 3nm And Beyond


Richard Gottscho, executive vice president and CTO of Lam Research, sat down with Semiconductor Engineering to talk about how to utilize more data from sensors in manufacturing equipment, the migration to new process nodes, and advancements in ALE and materials that could have a big impact on controlling costs. What follows are excerpts of that conversation. SE: As more sensors are added int... » read more

Engineering The Signal For GDDR6


DDR1 through DDR3 had their challenges, but speeds were below one gigabit and signal integrity (SI) challenges were more centered around static timing and running pseudo random binary sequence (PRBS) simulations. Now, with GDDR6, we are working on 16 to 20 gigabits per second (Gbps) signaling and even faster in the near future. As a result, engineering the signal for GDDR6 will require careful ... » read more

Latency Under Load: HBM2 vs. GDDR6


Steven Woo, Rambus fellow and distinguished inventor, explains why data traffic and bandwidth are critical to choosing the type of DRAM, options for improving traffic flow in different memory types, and how this works with multiple memory types.   Related Video GDDR6 - HBM2 Tradeoffs Why designers choose one memory type over another. Applications for each were clearly delineate... » read more

Power/Performance Bits: May 6


Compressing objects Computer scientists at MIT propose a way to improve data compression in memory by focusing on objects rather than cache lines. "The motivation was trying to come up with a new memory hierarchy that could do object-based compression, instead of cache-line compression, because that's how most modern programming languages manage data," said Po-An Tsai, a graduate student at... » read more

Comparing New Memory Types


After decades of research and development, three new types of memory—magnetic RAM (MRAM), phase change memory (PCRAM) and resistive RAM (ReRAM)—are moving toward commercial adoption, making this an exciting time for the semiconductor and computing industries. All three of these emerging memories are enabled by new materials and will require breakthroughs in process technology and manufactur... » read more

Finding The Bottom Of The Memory Trough


In announcing its Q2 fiscal 2019 results, Micron Technology, Inc. provided lower-than-expected revenue guidance of between $46 billion and $50 billion for the current quarter. However, what was particularly noteworthy was the company’s announcement to cut output by 5 percent due to weaker-than-expected market demand and its prediction that its customers’ inventory correction will last until... » read more

← Older posts Newer posts →