For IoT and Automotive Applications, Embedded MRAM Promises Cost-Effective and Low-Power Solution 

By David Lammers

One reason the International Electron Devices Meeting (IEDM) is an important event is to see how the semiconductor industry is converging on a technology option, be it hafnium-oxide gate oxides, immersion lithography, or, in this case, magnetic random-access memory (MRAM).

At the 2019 IEDM, held in December in San Francisco, the major foundries and Intel all presented MRAM technologies that can be embedded in CMOS logic devices. While it is fair to say GLOBALFOUNDRIES has an edge on the others in terms of reliability and manufacturing experience, the other companies have clearly embraced MRAM as well.

MRAM’s day has come largely because embedded NOR flash requires too many masks—a dozen or more—to manufacture at the 28nm node and beyond. Embedded NOR flash also requires a high-voltage capability to write data, and the write time is quite long. MRAM has its challenges, as well, but it is faster and less power-hungry than eFlash.

Big Power Savings

“If your applications write a lot to NOR flash, then you are going to love MRAM,” said Jim Handy, the veteran memory analyst at Objective Analysis, based in Los Gatos, California. “Flash consumes a lot of power, a phenomenal amount, because it takes so long to write and requires high voltages. If you move to MRAM, there is big power savings. The write power drops by a couple of orders of magnitude, while the read power of MRAM is about the same.” 

Handy makes the point that companies developing microcontrollers have a choice: they can either load up on SRAM for working memory and put the code storage on an external (discrete) NOR flash; or they can make the jump to embedded MRAM (eMRAM). Since SRAM requires six transistors to store a bit, MRAM typically has about double—or better—the density improvement, he said. 

Additionally, in systems where the SRAM requires battery backup, non-volatile MRAM is often more cost effective than the combined chip-plus-battery cost of embedded static RAM (SRAM), he said.

At the 2019 IEDM, an entire session was devoted to eMRAM. After presenting GF’s latest eMRAM reliability data, Vinayak Bharat Naik, the Singapore-based technical lead for GF’s embedded MRAM effort, said he welcomed having four companies—GF, followed by Intel, Samsung, and TSMC—pushing eMRAM at same time.

“For the customers, if they want to move to a new technology from a conventional technology that they have been using for a long time, it cannot be sudden,” Naik said. “Once an end customer starts up on MRAM, they will grow more confident in the idea of replacing conventional memory with MRAM.”

eMRAM Reliability and Manufacturability 

Over the past year, several clients have asked GF to share additional data showing its eMRAM technology could meet all reliability tests for production, as well as withstand strong external magnetic fields that might disturb stored data. 

GF’s 2019 IEDM presentation focused on providing an answer to these questions, and it was a positive story to tell. 

Naik’s IEDM paper showed the manufacturability of eMRAM on GF’s 22nm FD-SOI embedded platform using advanced magnetic tunnel junction (MTJ) stack/etch/integration processes by achieving a fully functional 40Mb macro at industrial operating temperature range, -40 to 125 degrees Celsius. It also showed the capability of meeting solder reflow requirements as well as robust product reliability with failure a rate of less than one part per million (ppm) at package level.

The magnetic immunity study showed the 40Mb eMRAM macro has the capability to withstand an extremely high magnetic field of 1,600 Oersteds in stand-by mode at 25 degrees Celsius, with failure rate less than 1 ppm for 20 min exposure. At 125 degrees Celsius, the failure rate was still less than one ppm at 700 Oe. Active-mode magnetic immunity—the capability of a chip to operate in the presence of a magnetic field—of 500 Oe was also demonstrated. Endurance remained excellent with failure rates less than 1 ppm up to one million cycles, with no degradation in resistance distributions after one million cycles, and no degradation during high-temperature operation at 500 hours. All of the results were with error correction (ECC) in off-mode.

“A magnetic field can be anywhere,” Naik said. “In the home, the charger for your phone, for example, can create a certain level of magnetic field. We need to make sure that both standby immunity and active-mode immunity are good so that the chip can operate as usual,” Naik said.

In 2018, at the major technology conferences including IEDM and the Symposium on VLSI Technology, GF demonstrated its eMRAM could withstand the solder reflow steps used in chip packaging, which would allow microcontrollers (MCUs) to be programmed prior to the package solder reflow steps. The JEDEC standard of five times solder reflow at 260 degrees Celsius for five minutes has been proven with package-level tests.

Improved Reliability Performance

At the 2019 IEDM, by showing eMRAM package level reliability data from all standard reliability tests and magnetic immunity, GF remains competitive in eMRAM technology, Naik said.

“At this IEDM, we showed that we are production-ready for industrial-grade applications, including wearables, internet of things (IoT), and many others,” he said. “GF has good production experience with 40nm and 28nm MRAMs, experience that carries over to the eMRAM market.”

GF engineers have continued to optimize the magnetic tunnel junction (MTJ) cell, including deposition and etch. “Over the past year, we improved both the MTJ stack and etch as well as integration processes to improve the endurance performance with better switching efficiency. And our yields were boosted to above the 90 percent level,” Naik said.

Saving on Energy Consumption

Tom Coughlin, a memory and storage consultant who served as general chairman of the annual Flash Memory Summit for 10 years, said eMRAM “has a lot of possibilities for embedded products at the edge or end points, especially those that are power-sensitive.”

The market for emerging memories, such as eMRAM, is positioned to take off, Coughlin said. “There is big growth in persistent networks, including Factory 4.0, which combine intelligent devices with AI for more-efficient factories. In addition, agriculture could be a big market, with more farmers placing productive wireless smart sensors in their fields. Also with health care there is a need for more efficient energy usage. Many markets will drive demand. And then there are things we haven’t even thought of yet, including many consumer applications, new uses for a fast energy-efficient memory are just starting to come on-line, but we haven’t recognized their potential yet.”

Naik said GF is taking it step-by-step, focusing first on IoT and industrial use, then automotive-grade eMRAM—where the temperature challenges are higher and where the data demands of autonomous driving require high-density on-chip memory—and then using MRAM as a level 4 cache, replacing some SRAM on processors.

And then there is another very large market, process in memory (PIM) computation, which was discussed often at the 2019 IEDM. PIM involves using some form of emerging memory in artificial intelligence (AI) computing. MRAM or other memory types, such as resistive RAM or phase-change RAM, could serve as the local processing element in edge devices. “Considering the superior performances of MRAM such as fast write speed, high endurance, high density, and low power, MRAM is unique among other NVMs and has a great potential for PIM computation for AI applications,” Naik said.

Process in Memory

Coughlin agreed about the potential of PIM. “Process-in-memory may be a bigger part of everything, putting AI applications in everything else,” he said. “We could do the training elsewhere, and have some learning capability on the device. At the very least, process-in-memory could run a model locally instead of at the data center.”

MRAM could also play a bigger role in data centers. “If the system is not using something, MRAM preserves the state, and when that data is needed it comes right back up. That takes us away from dependence on volatile memory toward a greater utilization of non-volatile memory. A lot of that today is driven by energy-sensitive applications, at edge points, but it could be used also in data centers,” Coughlin said.

Karim Arabi, CEO of San Diego-based Atlazo Inc., spoke at IEDM about change coming to edge devices. Autonomous driving is just one form of edge computing that will require “tons of data,” he said.

Advanced driver-assistance systems (ADAS) requires “low latency computing that is near the sensor,” Arabi said.

“When it comes to data aggregation and training, we can’t beat the cloud for computing power and data size. But other applications require much better power efficiency, and edge computing is 100 to 1,000 times less costly in terms of power than transmitting power over wireless links to the cloud. And for privacy reasons, a lot of data needs to stay local,” Arabi said.

In typical von Neumann architectures, about 75-95 percent of power is consumed by moving data between the memory and the processor. “With new memory architectures such as MRAM and PC-RAM, we can replace some SRAM with MRAM, and also move data from off-chip DRAM to on-chip MRAM. Either MRAM or PC-RAM could create a new paradigm in computing,” Arabi said. “Over the next 10 years, as neuromorphic computing takes hold, MRAM and PC-RAM will become even more key.”

New Compute Architectures

GF is positioning itself as a leader in MRAM, and embracing its potential for empowering GF clients to develop differentiated, feature-rich products, as well as to drive new technologies as potential new compute architectures.

Ted Letavic, chief technology officer and vice president for computing and wireless infrastructure at GF, said “we now have a connected society, and if you can’t process the data that we have within the power envelope, if can’t do the data analytics, then you can’t monetize, or even implement, AI. We have to be able to do the analytics, and that is either compute at the edge or in the data center.”

Moving forward, privacy will drive data to edge devices, where MRAM could play a role. “We all have personal data posted everywhere, from the edge to the data center. We would like to move that to the edge, to secure your data and be more private.”

A second factor driving edge computing is bandwidth. While 5G delivers more data to the data centers, that approach becomes impractical as the volume of mobile data accelerates. “Even with the huge promise of 5G, or even 6G, every bit that you have to transmit to the data center to compute takes bandwidth. We would like to get to the point where we have efficient compute engines at the edge. Then we could send the metadata—the result only—transmitting the result, not the raw data.”

Letavic said several major research centers are engaged with GF to explore these new approaches to edge computing.

“It is so much more than a silicon solution. We have to really change the compute architecture. Instead of just talking about new transistors and ways to handle electrons and photons, we are talking about new architectures,” Letavic said in an interview at the 2019 IEDM.

MRAM could play a major role in what Letavic calls the coming “renaissance of computer design.”

“For the first time in 30 years, we have opened the toolkit and are looking at non-Von Neumann architectures, where the power benefits are tremendous. We could achieve a 100 or 1,000 times lower power with dedicated architectures.”

Because the process-in-memory approach is so power efficient, MRAM could play a central role in these non-Von Neumann architectures. “As device technologists, we could keep improving the technology for the next 30 years, and we are still not going to get to a power point that meets our aspirations,” Letavic said. “We have to change the architectures and software stacks. New architectures bring new device types, new features on platforms, and new approaches to the compute problem.”

GLOBALFOUNDRIES’ first 10 years were marked by significant progress as well as some growing pains, as we chronicled in Part 1 of this series, and by a major pivot in corporate strategy, as detailed in Part 2. In this third and final installment, we’ll look at how GF, with a newfound sense of identity and purpose, is remaking itself to capitalize on the fundamental changes taking place in the semiconductor industry.

by Gary Dagastine

Many people from outside our industry call semiconductors “computer chips.” It’s a term that should perhaps be retired, in light of the fact that semiconductors are increasingly used not only for traditional computing, but in an exploding number and variety of applications in virtually every field of human endeavor where data can be put to good use. Wearable fitness devices, cars that are increasingly autonomous, and voice-activated personal assistants are just a few examples.

These diverse and fast-growing applications require semiconductor solutions tailored to their specific needs. Therefore it is of paramount importance to have the ability to integrate features such as 5G-ready radio frequency (RF) and mmWave capabilities, low power consumption, embedded non-volatile memory, high-voltage capabilities, silicon photonics, advanced packaging, and others, into cost-effective solutions.

As it begins the second decade of its corporate life, GLOBALFOUNDRIES is in an enviable position to serve such specialized requirements, with 15 node-based technology platforms, 14 unique and/or best-in-class application feature sets, and thousands of titles in its intellectual property (IP) ecosystem. Combined, these assets enable GF to offer literally tens of thousands of different application solutions. GF calls this formulation its Innovation Equation, and no other specialty foundry on the planet has anything close to it.

These offerings are being enhanced by the pivot away from extreme scaling that GF CEO Tom Caulfield initiated last year. The pivot has repositioned the company, by shifting resources away from 7nm technology development, and into more comprehensive and intense efforts to bring further innovations to the platforms already in its portfolio. In this way, GF is adding value for clients without introducing dramatically more costly manufacturing processes.

New Markets Bring New Opportunities

As part of its ongoing corporate transformation, GF recently created three strategic business units, each dedicated to serving specific high-growth semiconductor markets. GF defines them as Automotive, Industrial and Multi-market (AIM), Mobile and Wireless Infrastructure (MWI), and Computing and Wired Infrastructure (CWI).

These markets are being driven by megatrends such as the Internet of Things (IoT), cloud computing, artificial intelligence/machine learning (AI/ML), 5G communications, and the increased use of electronics in automotive systems, among others.

“As fast as these end-markets are growing now, we believe the opportunities for us in these segments will grow even faster as more and more devices are connected,” said Ted Letavic, Chief Technology Officer and Vice President of CWI at GF, who helped lead the development of all three new strategic business units. “Our specialized application solutions are both enabling and benefitting from these developments.”

GF estimates that about $47 billion of foundry business within these markets is addressable with solutions in the space where it plays – at or above the 12nm technology node. Of that total, the AIM business opportunities account for $24 billion, while MWI and CWI represent $15 billion and $8 billion, respectively.

The AIM market segment contains applications such as the IoT, which comprises networks of connected devices in our environment that sense, store, and transmit data. Automotive applications are also included in this segment, enabling features such as advanced driver assistance systems, automotive radar, powertrain control, and many others.

In the MWI market segment, a key driver is the advent of 5G wireless communications. Its lower latency and blazingly fast data transmission speeds are expected to enable universal mobile connectivity, dramatically lower data transmission costs for network operators, and spawn a host of new applications. It is estimated that as many as one trillion intelligent devices will be connected through 5G networks by 2035.

The CWI segment, meanwhile, is driven by the explosive growth in cloud computing and artificial intelligence/machine learning.

Innovative GF Solutions Enable New Applications

Innovation and success in these markets will be driven not by increasingly shrinking semiconductors, but by proven semiconductor platforms that are carefully designed and optimized to deliver specific features and performance.

“For the most part, 7-nanometer-and-beyond technology is irrelevant to these market opportunities, which is a key reason why GF turned its focus away from scaling,” said John Pellerin, GF’s Vice President of Global R&D Operations, and RF and SiPh Technology Solutions. “Other types of innovation that are tailored to specific application are required. We call these domain-specific solutions, and these sit squarely in GF’s wheelhouse.”

IoT applications are a case in point. Typical IoT devices might consist of a sensor with an analog interface, memory for code and data storage, RF capability for data communications, a processor to control the device and process the data, and very likely a battery and battery interface. Most of the time these devices may be in sleep mode, so ultra-low current leakage is a key requirement. Upon being awakened by a signal, however, the device must immediately switch into a higher-performance mode to fetch or store data in memory, process it, and then transmit or receive that data.

An expensive 7nm bulk CMOS logic chip doesn’t offer any practical advantage for handling these disparate functions. However, a system-on-chip (SoC) based on GF’s 22FDX® FD-SOI platform, with embedded MRAM memory and industry-leading RF capabilities, is an ideal solution. The native performance and power-efficiency of the 22FDX platform can be further enhanced with GF’s adaptive body-biasing feature to enable the system to be tuned dynamically, as needed, for higher performance or greater power efficiency. As a result, GF’s clients can more easily and cost-effectively develop novel IoT devices that sip power when quiescent, provide whatever high-performance capabilities might be required, and integrate other necessary features.

Cloud computing and AI inference/training in data centers provide another example of where node scaling may be less relevant, and where GF’s domain-specific solutions are critical enablers. For example, power consumption in data centers is a huge issue: in 2007 they consumed about 3 percent of total U.S. power output, but in 2020 it’s projected to increase to 7 percent. Clearly, every watt counts.

The biggest part of the AI power budget in data centers is moving data back and forth between memory and processors, which is why there is growing use of technologies such as compute-in-memory to reduce both power and latency. GF is developing many innovative solutions that address the issue. One is to reduce the power requirements and boost the performance of GF’s 12LP FinFET platform by adding features like very low voltage (0.5V) and dual-work function SRAM memories. Another is to use MRAM in an SRAM-like operating mode, the goal being to replace 6T SRAM with a memory that has three times the density at lower power.

Silicon photonics is another key enabler for lower latency data transmission and higher energy-efficiency within data centers. GF’s silicon photonics (SiPh) optical communication chips provide ways to significantly increase bandwidth to more easily transmit massive amounts of data, enabling new levels of performance.

Another GF innovation for data centers is 2.5D packaging technology. In one example, GF is working with Enflame Technology on an accelerator for fast, power-efficient cloud-based AI training platforms within data centers. The AI accelerator is based on GF’s 12LP FinFET platform and is co-packaged with other chips in an advanced 2.5D package, enabling Enflame to net the same performance they would get with a more highly scaled technology, but at a lower cost and with more flexibility in product functions.

Employees and Partners are Key

The transformation of GF into a company that is more aligned with the heart and the future of the semiconductor industry hasn’t taken place in a vacuum. GF’s thousands of employees bring technical expertise as well as dedication and commitment to their jobs, day in and day out, and the company wouldn’t be what it is without them.

In addition, GF has built deep collaborations and partnerships with networks of outside experts who provide the specialized knowledge and tools needed to ensure that GF’s clients can quickly, easily, and cost-effectively go from initial idea to finished, packaged, and tested product.

“Simply put, we are truly a collaborative foundry, and through our client enablement organization, we are open to working with any client in any way that best helps meet their objectives,” said Jim Blatchford, Vice President of Technology Enablement at GF. “In my previous roles at other companies, I was a customer of virtually every foundry in existence, so I know what it’s like to be on the other side of the desk. We understand what clients want, and we see our role as helping everyone, from big players who know exactly what they want from enablement, to smaller clients who need our guidance to tune their technology to address specific end markets.”

Accordingly, GF now has more than 100 ecosystem partners spanning IP, EDA, services such as design/technology co-optimization, and assembly and test. There are more than 3,400 IP titles across all of GF’s technology platforms, from more than 40 IP partners, with an additional 1,000 IP titles currently in active development. Over the past five years, more than 1,500 client designs have been enabled by these ecosystem partners.

GF also has put in place two partner ecosystems to help clients make the best use of key GF solutions. One is the FDXcelerator™ ecosystem to facilitate 22FDX SoC design and reduce time to market. The other is the RFwave™ partner program, an ecosystem of companies who have teamed with GF to help clients use GF’s various RF technology platforms to develop and bring differentiated solutions to market in less time by simplifying design.

Looking Ahead

GF’s first 10 years may have been occasionally tumultuous, and time will tell what the next decade holds in store. But as things stand, GF can look ahead knowing it has a realistic business strategy, a clear view of the market opportunities before it, innovative technical solutions that will ensure its clients’ success, and the human capital and external partners to provide all of the expertise and tools that clients are ever likely to need.