Partnership Puts ReRAM in SSDs

发布时间:2017-09-28 00:00
作者:Ameya360
来源:Gary Hilson
阅读量:1158

  Solid state drives (SSDs) are pretty much synonymous with NAND flash, but there have been attempts to use a different persistent memory with varying degrees of success.

  Mobiveil Inc. and Crossbar Inc. recently announced they are collaborating to use resistive random access memory (ReRAM) in an SSD. The collaboration will apply Mobiveil's NVMe SSD IP to Crossbar's ReRAM IP blocks. The goal is to enable 10 times more IOPs at one-tenth of the latencies of flash NVME SSDs to speed up access to frequently requested information in large data centers, the companies told EE Times in a joint telephone interview.

  Mobiveil CEO Ravi Thummarukudy said the company's NVMe, PCIe and DDR3/4 controllers can easily be adapted to accommodate the Crossbar ReRAM architecture, which is capable of six-million 512B IOPS below 10us latency. He said Mobiveil's NVM Express Controller architecture is designed to optimize link and throughput utilization, latency, reliability, power consumption and silicon footprint, and can be used along with its PCI Express (PCIe) controller and Crossbar's ReRAM controller.

  Mobiveil's NVMe controller IP is outfitted with an AXI interface that simplifies integration with FPGAs and SoCs. Other IP subsystem components include PCIe Gen 3.0, DDR3/4, and ONFI controllers. An FPGA development platform includes BSPs and drivers for validating the NVMe IP solution against user applications. Crossbar ReRAM technology, meanwhile, can be integrated in standard 40 nm CMOS logic or produced as a standalone memory chip.

  Thummarukudy said there's been a heightened level of interest in persistent memory since Intel and Micron introduced 3D Xpoint. Mobiveil and Crossbar have been working together for the last year, said Sylvain Dubois, vice president of strategic marketing and business development at Crossbar, and in addition to developing IP for ReRAM-based SSDs, they are also working on incorporating ReRAM into NV-DIMMs. “The NV-DIMM is the natural evolution of what we're doing with the SSD," Dubois said.

  The key benefit of using ReRAM in an SSD is that it reduces storage controller complexity by removing large portions of the background memory accesses needed for garbage collection. It also provides independent, atomic erasure by eliminating the need to build large-block memory arrays in flash designs.

  Neither Mobiveil nor Crossbar are building actual SSDs or NV-DIMMs. Rather, they have developed the IP so others can build their own solutions, said Thummarukudy.

  The Mobiveil/Crossbar collaboration is not the first attempt to make SSDs out of something other NAND flash, said Jim Handy, principal analyst at Objective Analysis, including failed attempts to make DRAM-based SSDs. This effort seems similar to Intel's Optane offering, he said. Handy added that Intel and Micron are adamant that Optane doesn't use phase change memory (PCM), considered a subset of ReRAM.

  Intel Optane is the only 3D Xpoint product to date, said Handy, and its chips are similar to that of Crossbar's ReRAM. “The chips themselves are both close to DRAM speeds but persistent," Handy said. "You put them behind an NVMe controller and its ends up being very fast."

  Handy said the biggest disappointment around 3D Xpoint is that Micron and Intel were promoting speeds a thousand times faster than that of NAND flash, but in reality it's only been seven or eight times faster.

  Handy expects Mobiveil and Crossbar's ReRAM-based SSD will suffer the same fate. “What's happening is the NVMe interface becomes a large part of the delay," he said.

  When Intel pushed the NVMe specification, it made sure to put in hooks that would support 3D Xpoint well before the company even announced it had 3D Xpoint. “That was definitely in the back of their minds," Handy said. "They weren't just optimizing an interface for NAND flash."

  Although there are customers who might want to build a ReRAM-based SSD, it's a small market, said Handy. “It's going to be pretty expensive," he added, saying its not dissimilar to the NV-DIMM market.

  “The current NV-DIMMS are more expensive than DRAM, they're way more expensive than SSDs, but offer blazing speed for people who want to pay for it," Handy said.

  Handy said the Crossbar ReRAM-based SSDs will find a niche with customers willing to pay top dollar for persistence and performance, adding that Intel is selling Optane at a loss because it helps the company sell more expensive processors.

  SSDs with 3D NAND are not in any danger, said Handy. “They will be far more economical than anything made of out Crossbar ReRAM," he said.

(备注:文章来源于网络,信息仅供参考,不代表本网站观点,如有侵权请联系删除!)

在线留言询价

相关阅读
Startup Demonstrates ReRAM Retention, Endurance
  An Israel-based semiconductor startup has reported positive results with its ReRAM technology.  Weebit Nano recently published preliminary evaluation results of endurance and data retention measurement on 4Kb arrays on 300nm cells. In a telephone interview with EE Times, CEO Coby Hanoch said the results successfully conclude the 300nm 4Kb characterization. The measurement was done under a variety of temperature and duration conditions at 150, 200 and 260 degrees Celsius, monitoring the ability of the ReRAM cells to maintain their resistivity levels within industry acceptable ranges.  Hanoch said 260 degrees Celsius is significant since it's the temperature used when soldering chipsets into printed circuit boards. Weebit Nano's 4Kb array kept its programmed data after 30 minutes at 260°C, exceeding the soldering requirement of 15 minutes at this temperature. This allows several soldering cycles. Data retention lifetime extrapolation showed the ability to keep written data for 10 years at above room temperature, he added, and high enough to meet the requirements of market segments such as industrial and automotive.  Other endurance characteristics were done under various voltage levels and timing durations to assess the ability of the memory to endure multiple re-write cycles. Hanoch said the endurance results were significantly higher than the program/erase cycling of existing flash technology. It's preliminary yet encouraging data that's critical for Weebit's ReRAM manufacturability and product reliability, and ultimately, commercialization. He said the company is on track to achieve its goal of a 40nm working cell by the end of the year.  Hanoch credits Weebit's collaboration with Leti for the fast pace of its ReRAM development because of it's strong knowledge and expertise in memory technology. The 4Kb array reliability results were achieved in Leti's pre-industrialization facilities in Grenoble, France, and were conducted in parallel to the Weebit's 40nm SiOx ReRAM cell development. “We've been making excellent progress really fast," he said. “We've done in two years what's taken others seven to 10."  Weebit's formation is atypical as well. The company was started in 2015 by non-technical people who licensed technology invented by Professor James Tour of Rice University. Tour is renowned for his work in the field of materials engineering and nanotechnology. Hanoch said Weebit has secured several patents to ensure commercial and legal protection for its technology, and Tour has demonstrated non-volatile memory behavior with SiOx, the most common and lowest cost material in the semiconductor industry. In the beginning, the company struggled to get funding, so it went to the Australian Stock Exchange to do a reverse merger to become a public company.  Hanoch said the differentiator for Weebit's ReRAM technology is it's based on standard materials using silicon oxide and same machines that exist in commercial fabs. Using standard tools and procedures should enable the company to reach productization faster while getting high yields at a lower cost. “There lots of companies developing ReRAM technology and next generation memories, but in the vast majority of cases they are trying to use materials not used in fabs," Hanoch  said.  Jim Handy, principal analyst with Objective Analysis, said the fact that Weebit is using standard materials is a big plus. “One of the great difficulties with other emerging memory technologies is that most of them use new materials that are either poorly understood compared to silicon, which the industry has been working with constantly for over 50 years, or they attack the silicon substrate," Handy said.  Weebit's ReRAM progress is not dissimilar to Everspin's advances with embedded MRAM in collaboration with GlobalFoundries announced earlier this year. The companies outlined eMRAM's ability to retain data through solder reflow at 260 degrees Celsius, and for more than 10 years at 125 degrees Celsius, plus read/write with outstanding endurance at 125 degrees Celsius, enabling it to be used for general purpose MCUs and automotive SOCs.
2017-11-23 00:00 阅读量:1038
ReRAM Goes 3D
Trade news

ReRAM Goes 3D

  Resistive random-access memories (ReRAMs) are a new breed of “universal” memory that could replace all other types, offering the speed of RAM but with the density and non-volatility of flash. To date, however, flash has managed to stay ahead of ReRAM by going 3D. Now the Moscow Institute of Physics and Technology (MIPT) says it has reengineered its ReRAM process to achieve a thin-film technique that is amenable to 3D stacking.  All ReRAMs work using memristors, in which migrating oxygen vacancies in the dielectric layer change the dielectric’s resistance to represent ones and zeros. In addition to MIPT, researchers from 4DS Memory Ltd., Crossbar Inc., HP Inc., Knowm Inc., and Rice University have created prototypes.  For 3D ReRAMs, “we needed not only to form oxygen vacancies in the dielectric layer, but also to detect them,” MIPT scientist Konstantin Egorov told EE Times. To do so, MIPT specialists used a method for observing the electron states in the bandgap of the dielectric that arise in the presence of oxygen vacancies.  “To study oxygen vacancies formed in the process of tantalum oxide film growth, we used an experimental cluster incorporating growth PEALD [plasma-enhanced atomic-layer deposition] and analytic XPS [X-ray photoelectron spectroscopy] chambers connected to each other by a vacuum tube with sample transfer manipulators. The cluster enabled us to grow and study deposited layers without breaking the vacuum,” Egorov said. “This is crucial because as soon as you take the experimental sample out of the vacuum, the nanolayer of dielectric oxidizes on its surface, which results in the annihilation of oxygen vacancies.”  Any semiconductor research lab could construct the unique ALD cluster by connecting the PEALD and XPS chambers and then adding robotic manipulators to transfer wafers between the chambers. For mass production, the cluster would not be needed, except to sample test wafers. A new assembly line, however, would have to be built that would compensate for the slow growth speed of ALD films.  If those efforts are successful, MIPT claims the resulting ReRAMs could be stacked vertically, yield a universal memory that would overcoming the limitations of 3D flash, which to date is restricted to 64 layers.  Deposition details  Though ALD is slow growing, it enables conformal coating of 3D structures, replacing the nanofilm deposition techniques used to date by MIPT and other research labs. The key difference is that ALD exposes a substrate, sequentially, to both a precursor material and a reactant material, and depends on the chemical reaction between the two to produce the active layer. MIPT’s technique also uses ligands—molecules chemically attached to a metallic precursor—to hasten the chemical reaction, but the ligands have to be removed before the active layer can be used in a device.  “Depositing oxygen-deficient films requires finding the correct reactants to both eliminate the ligands contained in the metallic precursor and control the coating’s oxygen content,” said Andrey Markeev, lead researcher at MIPT. “After much experimentation, we successfully used a tantalum precursor containing oxygen, and a plasma-activated hydrogen reactant.”  Next, the researchers plan to optimize the process and increase the ALD speed to enable mass production of 3D ReRAMs.  Funding for MIPT’s work was a provided by a Russian Science Foundation grant and MIPT’s 5-100 Program within the Russian Academic Excellence Project.
2017-07-03 00:00 阅读量:962
  • 一周热料
  • 紧缺物料秒杀
型号 品牌 询价
RB751G-40T2R ROHM Semiconductor
CDZVT2R20B ROHM Semiconductor
BD71847AMWV-E2 ROHM Semiconductor
MC33074DR2G onsemi
TL431ACLPR Texas Instruments
型号 品牌 抢购
STM32F429IGT6 STMicroelectronics
IPZ40N04S5L4R8ATMA1 Infineon Technologies
ESR03EZPJ151 ROHM Semiconductor
BP3621 ROHM Semiconductor
TPS63050YFFR Texas Instruments
BU33JA2MNVX-CTL ROHM Semiconductor
热门标签
ROHM
Aavid
Averlogic
开发板
SUSUMU
NXP
PCB
传感器
半导体
相关百科
关于我们
AMEYA360微信服务号 AMEYA360微信服务号
AMEYA360商城(www.ameya360.com)上线于2011年,现 有超过3500家优质供应商,收录600万种产品型号数据,100 多万种元器件库存可供选购,产品覆盖MCU+存储器+电源芯 片+IGBT+MOS管+运放+射频蓝牙+传感器+电阻电容电感+ 连接器等多个领域,平台主营业务涵盖电子元器件现货销售、 BOM配单及提供产品配套资料等,为广大客户提供一站式购 销服务。