From Blue to Orange: One Year Later June 29, 2016When I sat down next to a yield engineer at lunch during a recent semiconductor manufacturing conference, we started talking. He said he worked at the Malta fab of GLOBALFOUNDRIES but had spent most of his career at IBM’s Fishkill fab, coming over last July 1 when GF formally acquired IBM Microelectronics. I asked him about the transition, whether the IBMers – especially those with some years at IBM — had felt anxieties about moving into a new organization. “Actually, it was the opposite,” he replied quickly. “At IBM there was a feeling that semiconductor manufacturing was no longer central to the organization’s key purpose. That went on for some time. So when we moved to GF, it was like ‘now we know what we are doing.’ We were finally in an organization where making semiconductors is the main purpose.” Some of the top managers in New York and Vermont express similar sentiments about a transition process that involved 5,000 people and two fabs. Geoff Akiki now runs the mask organization at GF’s Burlington, Vt. site, but earlier he was an Integration Executive tasked with bringing IBM Microelectronics into the GF fold. The integration team created a simple color code: Red for people within IBM Microelectronics, Orange for GF people, and Blue for those who would remain with IBM, largely to manage the foundry relationship with GF. “There was some trepidation, by both IBMers and GF people. Some of the GF people wondered how this would affect their roles in the expanded organization. For the most part, the Red people were hugely excited, because they felt that GF was doing chip manufacturing for a living and had strategies that were better than what they had seen for 10 years.” If GF brought resources and focus, IBM brought world-class engineers, an ASIC business, and radio frequency (RF) technologies, based on both RF-SOI and SiGe processes, a valuable asset in this era of mobility. Tom Caulfield, general manager of Fab 8 in Malta, N.Y., said about 600 people came to Malta after the formal integration date of July 1, 2015. However, prior to that a fairly large number of engineers had left IBM, seeing the handwriting on the wall, to apply for positions at GF. When Caulfield took on his current job in 2014, he hired what he calls “A players” from various companies, including Intel and Samsung, and large numbers of experienced IBMers. “At IBM, microelectronics was seen as a cost center, a means to an end, but not The Business. These guys were chomping at the bit to go where making semiconductors is the real game.” Bringing 5,000 people – about 3,000 from the Burlington fab and another 2,000 who worked at Fishkill, N.Y. – into GF was a large task. To prepare, Akiki said the integration team created 14 work streams, and organized meetings where as many as 200 people would participate. Two key decisions eased the transition process. The most important was that there would be essentially no layoffs. GF CEO Sanjay Jha had laid down a philosophy of ‘disturb as little as possible’ during the integration. Bringing essentially everyone over from IBM reduced anxiety levels considerably, Akiki said. Secondly, GF decided to acquire the two IBM manufacturing sites (and the facilities support staff) at Burlington and Fishkill in their entirety, obviating any need to partition the properties. (IBM kept the packaging technology center in Bromont, Canada). IP Integration At the recent 2016 Design Automation Conference (DAC) in Austin, one speaker noted that when Company A acquires Company B, typically the first to be laid off are the EDA engineers at Company B. With them goes the historical knowledge of what intellectual property (IP) is owned by Company B, and how those IP cores can be used by the design teams. Akiki said from the outset intellectual property was seen as “a large part of the value” of the IBM integration deal. “We bragged about the number of patents we picked up. But we knew that IP often languishes, and we set a specific objective to integrate both the people and the IP files.” A major effort went into making sure that all of the IP had been categorized, so that GF could “eventually declare we have received the key deliverables.” Gary Patton became the chief technology officer at GF as part of the integration, and it is probably fair to say that his technology development organization has benefited the most from the deal. One consequence of Malta being a “green field” site, Caulfield noted, is that it takes time to build up the talent. That was especially true in technology development (TD), with Patton adding that maintaining a TD schedule “has been a key knock” in the past. For the 10/7nm development program, Patton said more than 50 percent of the people are from IBM, adding that the 10/7nm development team draws on experience from the 14nm team. Caulfield emphasized that Malta has benefited from the “added scale” of ASIC business that came to GF as part of the IBM deal. IBM engineers had developed high-performance ASIC cores, such as a 56 gigabit/s Serdes core, that Caulfield said are not available from other foundries. “We are refreshing our ASICs platform at 14nm, and I can tell you that I entertain about one new customer at week at this site, with at least half of them being ASIC customers.” I asked about the challenges which remain, a year after the initial integration. The most important challenge is to further knit together the two human organizations, making sure that the people with two or three decades of service during their careers at IBM share their knowledge with the engineers at Malta, and vice versa. As Caulfield said, “Our job is to get more intermixing going on, to get more balance. We don’t want to bring in all of these great technologists and then not leverage their skills.” Patton, who earlier managed IBM’s Semiconductor Research and Development Center (SRDC), promises to take advantage of the talent he now manages at GF. “In the 10 years I led the SRDC, I can tell you we always had more performance than Intel. The industry has been heavily focused on mobile, and now customers are not happy with the performance improvements they are seeing from our competitors, compared with 16nm. They are looking for more performance. And guess what? We just brought in a specialty team here at GF that knows how to do performance.”
针对工业物联网、汽车自动化和户内导航应用，A*STAR IME将加强MEMS技术的能力。 June 8, 2016A*STAR IME’s collaborative partnership with industry will enable the development of cutting-edge industrial-grade sensors to heighten performance and achieve cost-effectiveness for MEMS devices
A*STAR IME’S Consortium To Deepen Capabilities In MEMS Technologies For Industrial Internet Of Things, Automotive, and Indoor Navigation Applications June 8, 2016A*STAR IME’s collaborative partnership with industry will enable the development of cutting-edge industrial-grade sensors to heighten performance and achieve cost-effectiveness for MEMS devices
GLOBALFOUNDRIES to Expand Presence in China with 300mm Fab in Chongqing May 31, 2016Company plans new manufacturing facility and additional design capabilities to serve customers in China Santa Clara, Calif., May 31, 2016 – GLOBALFOUNDRIES today announced the signing of a memorandum of understanding to drive its next phase of growth in China. Through a joint venture with the government of Chongqing, the company plans to expand its global manufacturing footprint by establishing a 300mm fab in China. GF is also investing in expanding design support capabilities to better serve customers across the country. “China is the fastest growing semiconductor market, with more than half of the world’s semiconductor consumption and a growing ecosystem of fabless companies competing on a global scale,” said GF CEO Sanjay Jha. “We are pleased to partner with the Chongqing leadership to expand our investment in support of our growing Chinese customer base.” The initial plan of the project includes upgrading an existing semiconductor fab to accommodate the manufacturing of 300mm wafers using GF’s production-proven technologies from its Singapore site. The proposed joint venture will provide immediate access to a state-of-the-art facility, accelerating time-to-market with production planned for 2017. “In recent years, Chongqing has followed the cluster model to vigorously develop the electronic information industry, becoming one of China’s most important locations for intelligent end products manufacturing,” said Huang Qifan, Mayor of Chongqing. “During the period of China’s thirteenth five-year plan, Chongqing will continue to develop the intelligent IC and other strategic emerging industries, and promote sustained and healthy economic development in the region. GF is a world-famous IC manufacturing company, and we welcome them to participate through cooperation to achieve mutual benefit and win-win. Cooperation between the two parties will help to enhance the production of intelligent IC technology in Chongqing, further improving the electronic information supply chain in Chongqing and the rest of China.” GF continues to strengthen its sales, support, and design services offerings in China, doubling over the past year with plans for continued growth. The company’s current presence is anchored by world-class design centers in Beijing and Shanghai, which have extensive expertise in custom designs supporting a robust ASIC platform, coupled with foundry design capabilities for a variety of technology nodes. These capabilities are complemented by key regional partners in its design and IP ecosystem. ABOUT GF GF is the world’s first full-service semiconductor foundry with a truly global footprint. Launched in March 2009, the company has quickly achieved scale as one of the largest foundries in the world, providing a unique combination of advanced technology and manufacturing to more than 250 customers. With operations in Singapore, Germany and the United States, GF is the only foundry that offers the flexibility and security of manufacturing centers spanning three continents. The company’s 300mm fabs and 200mm fabs provide the full range of process technologies from mainstream to the leading edge. This global manufacturing footprint is supported by major facilities for research, development and design enablement located near hubs of semiconductor activity in the United States, Europe and Asia. GF is owned by Mubadala Development Company. For more information, visit https://www.globalfoundries.com. Contacts: Jason GorssGF(518) 305-9022[email protected]
Executive Perspective: The Big Shrink May 27, 2016By Gregg Bartlett When we in the semiconductor industry talk about shrinking, most of the time we mean device scaling. But another type of shrinking is taking place, too, and I want to talk about its implications for the industry. I’m referring to the shrinking number of companies at the top of the market, resulting from the continuing consolidation of fabless companies. Since the beginning of 2014, mergers and acquisitions valued at more than $150 billion have taken place in our industry, nearly 10 times the yearly average. These have been driven by a combination of low interest rates, saturation in the mobility space, slowing growth rates and an overall squeeze on profitability This wave of corporate tie-ups is leaving a huge amount of disruption in its wake for foundries and foundry customers alike, as newly integrated companies seek to exercise their larger-scale purchasing power, consolidate their supply chains, simplify their roadmaps and drive a multitude of new integration synergies. Disruption is also occurring because of new architectures and more complex packaging technologies, which are making it harder to distinguish where wafer processing ends and packaging begins. As a result, we are seeing a growing trend by system houses to engage directly with foundries for complete turnkey solutions, from fab to test to packaging to finished goods inventory – these may even involve design centers along with the complete supply chain. Foundries face a number of critical requirements and challenges as a result of these trends. One is that large-scale foundry operations are more important than ever, as the largest customers simplify their supply chains and ask their foundry partners to do more. The fact that margins are under great pressure also drives the need for scale, and for continuing cost reductions. Moreover, decision-making has become a higher-stakes process than ever before as a result of the fewer number of customers and foundries at the leading edge. GLOBALFOUNDRIES is pursuing a multi-faceted approach to meet these challenges. With regard to scale, we are building the capability in our Dresden fab for hundreds of thousands of 22FDX® FD-SOI wafer starts annually, in a facility that has the ultimate capacity for more than a million wafer starts a year overall. Also, the acquisition of IBM’s semiconductor operations has given us more capacity, and more supply for customers. With the two additional fabs – one in Burlington, Vermont and the other in East Fishkill, New York – we can expand our capacity in RF SOI as well as other processes. And, on the ASIC side we have a very strong 14nm ASIC business and IP portfolio, which directly connects our foundry to end-market system houses. Meanwhile, in collaboration with select design partners, equipment and material suppliers and OSAT partners, our offerings span the range of design-fab-turnkey solutions. And I’m proud to say there is no better example of our end-market expertise – that is, our ability to engage architects – than our work to ensure that viable, cost-effective solutions exist for the forthcoming move to 5G cellular networks. In the end, the industry is changing so swiftly and deeply that nobody can be certain how it will evolve. But one thing is for certain: We at GF are planning and implementing solutions that address as broad a swath of our customers’ current and future requirements as needed, regardless of how things may evolve.
格芯发布性能改进的130nmSiGe射频技术以推动下一代无线网络通信 May 23, 2016优化的SiGe8XP技术将为大量的RF应用带来低成本、高性能的毫米波20GHz产品 加州圣克拉拉，2016年5月23日— 格芯今天公布了下一代的射频硅设计方案，并将其添加到其矽鍺高性能技术组合当中。此技术在多方面进一步增强了性能，如行车雷达、卫星通信、5G毫米波基站和其他有线或无线通信应用。 格芯的SiGe 8XP技术是公司130纳米高性能SiGe类别里的最新拓展的技术，使客户可以拥有开发更快的流率、更远的距离以及更少功耗的射频方案的能力。对比起它的上一代产品SiGe 8HP，此项先进技术提供改进的异质结双极晶体管性能、低噪音系数、高信号完整性和高达25%的最大震荡频率（增幅至340GHz）。 毫米波频率波段下操作的高带宽通信系统对复杂性和性能的高要求，创造了对高性能硅方案需要，也创造了对高性能SiGe在射频前端5G智能手机和其他毫米波相位阵列消费者应用上的机会。这些应用依赖于SiGe的优异表现，包括通讯基站基建、回程线路、微信和光纤网络。 “5G网络将会对RF SOC设计带来另一个级别的革新，支持高带宽数据传送并达到增加数据传输率和低延迟应用的要求。” 格芯射频业务部高级副总裁Bami Bastani博士说道，“格芯的SiGe 8HP和8XP技术提供性能、功耗、效率之间的出色稳定性，是客户可以对下一代移动和基建类硬件开发独特的射频方案。” “格芯的SiGe技术领导地位和综合性的PDK帮助我们的设计者可以更快的进行优化开发，并更好的提供差异化的毫米波解决方案。”Anokiwave 总裁Robert Donahue说道，“SiGe 8XP使我们将性能带到更高的级别，其面向未来的毫米波方案专为供应商准备，让他们可以市场需求更领先一步，以满足稳定的连接和并处理爆炸式增长的移动数据流量。” 当未来的5G部署持续刺激着基站和小面积单元的流行，SiGe 8HP 和8XP为毫米波频率独特方案的价值、功率输出、效率、低噪音和线性提供了平衡，并可以应用于下一代移动基建硬件和智能手机射频前端。格芯的SiGe 8HP 和8XP高性能产品使芯片设计者可以发展比矽鍺更经济、比CMOS更高性能的技术同时，集成重要的数字与射频功能。 除了在毫米波频率高效操作高性能晶体管外，SiGe 8HP 和8XP引进了可以减小裸晶尺寸增加面积利用率的技术革新。全新的铜金属化功能提供改进的电流传导能力以及在100度时5倍的电流密度，或者，同样电流密度下高于标准铜线25度的操作温度。此外，格芯已经可以提供已通过生产验证的硅通孔技术。 SiGe 8XP设计套装已经为您准备好了。欲了解更多详情关于格芯130纳米SiGe高性能技术方案，欢迎在国际微博论坛活动时参观1443号展台，活动时间5月22到27日，位于加州旧金山，或请登录网址 https://www.globalfoundries.com. Erica McGill GF (518) 305-5978 [email protected]
GLOBALFOUNDRIES Releases Performance-Enhanced 130nm SiGe RF Technology to Advance Next Generation Wireless Network Communications May 23, 2016Optimized SiGe 8XP technology will enable low cost, high-performance mmWave 20 GHz products for a broad range of RF applications Santa Clara, Calif., May 23, 2016 – GLOBALFOUNDRIES today announced a next-generation radio-frequency (RF) silicon solution for its Silicon Germanium (SiGe) high-performance technology portfolio. The technology is optimized for customers who need improved performance solutions for automotive radar, satellite communications, 5G millimeter-wave base stations and other wireless and wireline communication network applications. GF’s SiGe 8XP technology is the latest extension to the company’s 130nm high-performance SiGe family and enables customers to develop RF solutions that deliver even faster data throughput, over greater distances, while consuming less power. The advanced technology offers an improved heterojunction bipolar transistor (HBT) performance with lower noise figure, higher signal integrity, and up to a 25 percent increase in maximum oscillation frequency (fMAX) to 340GHz compared to its predecessor, SiGe 8HP. The complexity and performance demands of high bandwidth communication systems operating in the mmWave frequency bands have created the need for higher performance silicon solutions. This creates opportunities for high-performance SiGe solutions in the RF front end of 5G smartphones and other mmWave phased array consumer applications in addition to the current applications that depend on SiGe for high performance, such as the communications infrastructure base stations, backhaul, satellite and fiber optic networks. “5G networks promise to bring a new level of innovation to RF SOC design to support high bandwidth data delivery and meet the demands for increased data rates and low latency applications,” said Dr. Bami Bastani, senior vice president of GF RF business unit. “GF’s SiGe 8HP and 8XP technologies offer an outstanding balance of performance, power, and efficiency that enable customers to develop differentiated RF solutions in next-generation mobile and infrastructure hardware.” “GF’s SiGe technology leadership and comprehensive PDKs enable our designers to develop performance-optimized, differentiated millimeter wave solutions quickly,” said Robert Donahue, Anokiwave CEO. “Utilizing SiGe 8XP allows us to take performance to even higher levels in future-ready mmWave solutions designed to help providers stay ahead of the demands for reliable connectivity, from anywhere, while handling exploding volumes of mobile data traffic.” With tomorrow’s 5G deployments poised to drive a proliferation of base stations with smaller cell areas, SiGe 8HP and 8XP are designed to help offer a balance of value, power output, efficiency, low noise, and linearity at microwave and millimeter-wave frequencies for differentiated RF solutions in next-generation mobile infrastructure hardware and smartphone RF front ends. GF’s SiGe 8HP and 8XP high-performance offerings enable chip designers to integrate significant digital and RF functionality while exploiting a more economical silicon technology base compared to gallium arsenide (GaAs) and higher performance than CMOS. In addition to high performance transistors for efficient operation at mmWave frequencies, SiGe8HP and 8XP introduce technology innovations that can reduce the die size and enable area-efficient solutions. A new Cu metallization feature provides improved current carrying capabilities with five times the current density at a 100C, or up to 25 degrees C higher operating temperature at the same current density compared to standard Cu lines. In addition, GF’s production-proven through-silicon-via (TSV) interconnect technology is available SiGe 8XP design kits are available now. For more information on GF’s 130nm SiGe high-performance technology solutions, visit our booth #1443 at the International Microwave Symposium from May 22-27, in San Francisco, California, or go online at globalfoundries.com/SiGe. About GF GF is the world’s first full-service semiconductor foundry with a truly global footprint. Launched in March 2009, the company has quickly achieved scale as one of the largest foundries in the world, providing a unique combination of advanced technology and manufacturing to more than 250 customers. With operations in Singapore, Germany and the United States, GF is the only foundry that offers the flexibility and security of manufacturing centers spanning three continents. The company’s 300mm fabs and 200mm fabs provide the full range of process technologies from mainstream to the leading edge. This global manufacturing footprint is supported by major facilities for research, development and design enablement located near hubs of semiconductor activity in the United States, Europe and Asia. GF is owned by Mubadala Development Company. For more information, visit https://www.globalfoundries.com. Contacts: Erica McGillGF(518) 305-5978[email protected]
RF Driving Next-Gen Processes May 16, 2016By Dave Lammers GLOBALFOUNDRIES is expanding its RF capabilities in two important ways: moving RF SOI manufacturing to larger wafers and a new technology platform at its East Fishkill 300mm fab. Secondly, RF IP development plays a key role in the 22FDX® platform.</em> Successful semiconductor companies are facing an interesting challenge: they must be able to create solutions for the fast-growing automotive and Internet of Things markets by combining three technologies that, historically, often have been separated: processors and other digital cores; memories, and RF. Subramani Kengeri, vice president of the CMOS Platforms business unit at GF, said “going forward, the SoCs for every emerging market will have a radio. With the 22FDX platform we have a cost-effective solution, with RF and analog on the same technology as digital.” RF and Digital Convergence Fully depleted SOI has advantages for on-chip RF. The planar transistors in the 22FDX technology have less variability than finFETs on a bulk silicon substrate, where the ability to control both the height and width of the fin is challenging. “FD-SOI technology provides better transistor-matching characteristics. Because 22FDX is planar, with much lower variability, that helps in building cleaner RF and analog alongside high-performance digital,” Kengeri said. To accelerate the 22FDX rollout, GF contracted with INVECAS, a Santa Clara-based IP vendor, for 22nm libraries and higher-level IP offerings that are exclusive to GF silicon. Other eco-system partners are engaged in the development of silicon-proven WiFi and Bluetooth cores as a priority. “Over 45 customers are in various stages of engagement and lead customers have taped out test chips. All the top five EDA providers have announced support for 22FDX. We are on track to qualifying the technology later this year,” Kengeri said, with high-volume production to quickly follow. RF SOI Moving to 300mm GF also is working on developing its next-generation RF SOI process, which continues its leadership in RF front-end silicon technologies. The foundry recently reached an important milestone in shipping its 20 billionth RF SOI chip. As the performance requirements for RF SOI technologies become more challenging and the demand continues to increase as a function of smartphone radio complexity, GF is working to address the next wave of innovation in mobile RF fronts. Enabling manufacturing on 300mm wafers is an important component of that strategy. Peter Rabbeni, senior director of the RF business unit, said “we have already proven that 300mm can provide a number of additional benefits besides capacity enhancement. The availability of new materials and smaller lithography are some of the capabilities of 300mm manufacturing that benefit device performance.” One of the key differentiators of RF SOI is that circuits are built on an engineered substrate — much different than the SOI substrate used for digital applications such as lowpower microcontrollers — which has characteristics that are better suited for high-performance RF. The substrate characteristics support the high isolation and low harmonic response needed in RF front end circuits, preventing radio interference and preserving signal fidelity, Rabbeni explained. GF worked closely together with substrate suppliers to develop an RF SOI technology which meets the stringent harmonic and linearity requirements that today’s RF front end switches and tuners need. LTE communications and carrier aggregation require next-generation RF SOI with improved insertion loss and linearity. Carrier aggregation, for example, introduces data rate expansion methods which binds two or more carriers to a single data stream. This introduces certain complexities in the RF path that need to be accounted for to insure any nonlinear products that are generated by this operation are minimized. “Another key trend we are observing is the integration of more digital content. There is significant adoption of the MIPI interface for RF front end control, for example, and this is now becoming a larger percentage of the die,” he said. Beyond the needs of the LTE standard, Rabbeni said the next-generation RF SOI process will set the foundation for the 5G cellular standard. Though the final 5G standard has yet to be ratified, customers are already developing 5G demonstration systems to intersect the 2018 and 2020 Olympics. Millimeter wave frequency operation seems best suited to deliver on the promises of 5G, including low latency, spectral efficiency, and high cell edge data rates. “If this is the direction the industry takes, more integration will be required than today’s RF SOI technologies can achieve,” Rabbeni said. Customers may integrate the beam formers, power amplifiers, phase-shifters, LNA’s (low noise amplifiers), and even some portions of the transceiver into a single chip under high-speed digital control. “For certain, going forward there is an expectation our customers will want significantly more integration with RF SOI. We are leveraging much of the learning that has been achieved on our 45SOI technology to help make this leap forward,” he said. In the end, it all comes down to the enablement which helps designers get to market quickly with their product. “We spend significant effort and take a lot of pride in providing very accurate models and high quality process design kits (PDKs), so that customers can be confident that what they simulate is exactly how the silicon performs when it comes out of the fab. We have decades of manufacturing experience in RF silicon technologies. There are not many large foundries that can make that claim.” These growing market opportunities have led to “a big focus on the transformation of our manufacturing capacity. We have made some very focused decisions on capacity additions for RF SOI and silicon germanium to make sure we can meet the anticipated demand. Expansion to meet the coming demand fromChina is a big focus for us,” Rabbeni said. As China’s cellphone users move to 4G- and LTE-capable handsets, and as the 5G standard begins to take hold, the demand for RF SOI and SiGe-based chips could expand quickly, just as it did several years ago. Watching the 300mm Move Joanne Itow, managing director at Semico Research (Phoenix), said she is watching closely to see how the transition to 300mm RF SOI wafers works out at GF and other foundry vendors. When IBM’s Burlington operation developed a silicon-based path to RF front end ICs, “the switch to RF SOI and away from GaAs was pretty quick, as the benefits were obvious. The foundries that have the ability to move to 300mm wafers have a leg up. Just having that option is a real plus,” Itow said. Itow said she is watching to see how the SOI wafer suppliers, primarily Soitec (Grenoble, France), respond with a reliable supply of 300mm RF SOI wafers, and how the foundries and customers take advantage of the larger wafer sizes. “We are looking at the next move to bringing products on to 300mm capacity. What we are being told by the foundries sounds good, and it sounds as if GF is in the right place, getting ready for the right markets. Now we will have to wait and see if it will work out or not,” she said.
To Bias or Not to Bias, That Is the Question May 12, 2016By Joerg Winkler One of the essential building blocks of applications in mobile, pervasive and intelligent computing space is a high-performance, low-power processor. For these applications, GLOBALFOUNDRIES 22FDX® platform with 22nm fully depleted silicon-on-insulator (FD-SOI) technology offers an optimal combination of performance, low power and cost. One big advantage of 22FDX is the ability to optimize performance and power by applying forward and reverse body bias to the transistors. The challenge for our design team was to successfully apply body-bias to enhance PPA of a quad-core ARM Cortex-A17 processor implemented in 22FDX FD-SOI technology. In GF’s webinar, Implementing an ARM® Cortex®-A17 Processor in 22FDX Technology, we examine a digital implementation flow with industry-standard EDA tools, the application of body-bias for specific design intents and power scenarios, provide analysis of physical architecture details and initial PPA results of an ARM Cortex sub-module. The concept of an optimizable technology platform holds great potential, but adopting a new platform often means adopting a new design flow as well. And engineers know that with new design flows, the road from concept to reality can be bumpy unless the implementation details are well thought out. Fortunately, the GF 22FDX FD-SOI design flow is architected to be very similar to the existing bulk flow. With support from all of the major EDA vendors, the 22FDX flow uses various design techniques (implant-aware, source/drain-aware, double patterning, UPF support) which have been deployed on earlier nodes. This case uses the Cadence tool suite from initial design creation to signoff. We detail the implementation of an ARM Cortex processor as a reference design, highlighting how to obtain a wide range of PPA results by applying both forward and reverse body bias to different domains in a floorplan. With this easily tunable tradeoff, you can effectively balance between higher performance and lower power to meet the overall performance specs and power budget of a SoC design. GF design IP for the ARM Cortex-A17 processor includes standard cell base libraries, power management kit and cache memory kit, each with support for body-biasing. The 22FDX platform is ready to adopt for new designs, with the starter kit of 22FDX digital design flow available now. To replay the webinar, click here. More information including videos and white papers are available at GF.com/22FDX.
FD-SOI: An Enabler of Disruption April 21, 2016 For years, Dan Hutcheson has stayed on the sidelines as the industry buzz has grown around FD-SOI technology. “I’ve been pretty quiet, because I never bought the cost argument. I never thought the decision would be swung by a couple of mask layers,” said Hutcheson, who is the CEO and Chairman of analyst firm VLSI Research. “But last year when I saw 22FDX® from GLOBALFOUNDRIES, I saw for the first time some game-changing features. Power didn’t matter five years ago, but now it’s a very power-stingy world. Designers are differentiating on power, not necessarily performance. The real-time tradeoffs in power offered by FD-SOI began to look pretty exciting to me.” Hutcheson wanted to get validation for his increasing interest in FD-SOI, so with the help of GF, he set out to conduct a survey of key influencers and decision-makers in the chip design ecosystem. In the following video, Hutcheson presents the key findings from his in-depth interviews, where he asked participants about some of the key technical and business reasons to design with FD-SOI, how the technology is positioned alongside industry FinFET offerings, and other questions designed to answer the overarching question: “Is FD-SOI disruptive, or just another process?” The answer? “No. It’s not disruptive, but it’s an enabler of disruption,” Hutcheson concluded. “The Internet of Things (IoT) is the most disruptive force out there. It will be as disruptive as the smartphone, and I believe that FD-SOI technology will be a critical enabler of this disruption.” To view Dan Hutcheson’s FD-SOI presentation, click here.
