Director Technology, Strategy
Vijay Rao is the Director of Technology and Strategy at Facebook. His responsibilities include defining and optimizing Facebook’s infrastructure while working closely with its software teams to design high-performing systems. He works closely with partners to define future products in their roadmap. He is passionate about all things related to scale out infrastructure. Before joining Facebook, he was a Technologist in the Office of the CTO at AMD, where he was primarily responsible for cloud-related products. He has also been the founder of an enterprise security company and a CPU design engineer at Intel. He holds four patents and has published several articles. He earned an MSEE from Purdue University.
Abstract: How We Use Flash at Facebook: Tiered Solid State Storage
Facebook obviously faces huge infrastructure challenges. Its 1.6 billion worldwide users create huge amounts of new data every day (including photographs and videos). Future demands will be even greater as Facebook enables richer experiences, increasing demands on storage and computing power. Facebook has been using flash memory for many years in a variety of ways, including for its databases. Disaggregation of resources has been an important theme at Facebook. This talk will describe how Facebook uses Lightning, their JBOF (Just a Bunch of Flash) to enable hierarchical storage. Furthermore, Facebook has been investigating the use of dense, ultra-low-endurance flash as a repository for large data items. Facebook expects to use new memory technologies such as Intel/Micron 3D XPoint™ to create new architectures based on persistent memory (non-volatile memory with DRAM-like access speeds).
Facebook builds products that enable people to connect and share through mobile devices and personal computers. The Company enables people to share their opinions, ideas, photos and videos, and other activities More information is available at Facebook.
VP/GM, Storage Business Unit
Darren Thomas is Vice President and General Manager of Micron’s Storage Business Unit, where he oversees the solid state storage business that ranges from hard disk drive replacements with solid state drives (SSDs) to enterprise-class storage solutions. He was previously VP/GM Storage at Dell, where he transformed the company from providing strictly OEM solutions to having its own distinct strategy and product lines, leading several acquisitions including EqualLogic and Compellent. He has also been VP/GM Storage at Compaq Computer, where he built the storage business to include server-based storage, midrange, and enterprise product lines. He has over 30 years of technical industry experience. An engineering graduate of the University of Memphis, Thomas was also instrumental in founding the Storage Networking Industry Association (SNIA).
Abstract: Turning Big Data into Critical Business Advantage with Flash Memory
Enterprises today have more data than they can currently handle with an unimaginable 2.5 quintillion bytes being produced daily. How can flash memory help them get the most business advantage out of their data and stay competitive? There are currently many technologies that can play important roles. 3D NAND provides faster, larger devices at lower cost. 3D Xpoint™ offers nonvolatile devices that behave more like standard memories, producing storage at memory speeds. NVDIMM makes standard packages act like nonvolatile memory. NVMe speeds up today’s SSDs by putting them on the well-established, high-throughput PCIe bus. Storage and system engineers must use all these approaches to design platforms that can handle more business critical data faster and produce more insight into every aspect of enterprise operations. They can also meet other pressing needs such as lower energy consumption and reduced footprint. The end result is that flash will play a critical role in keeping enterprises both innovative and competitive in a fast-moving worldwide environment.
Micron is one of the world's leading providers of advanced semiconductor memory solutions. Micron’s DRAM, NAND and NOR flash components are used in today’s most advanced computing, networking, storage, and communications products, including computers, workstations, servers, storage arrays, cellphones, wireless devices, digital cameras, gaming systems, and many embedded applications. More information is available at Micron.
Shigeo (Jeff) Ohshima
SSD Technology Executive
Storage & Electronic Devices Solutions Company
Storage & Electronic Devices Solutions Company
Yoichiro Tanaka is a leading strategic researcher into next-generation medical data center initiatives. His research activities focus on new storage and computer architectures, with an emphasis on genomic and brain neurosciences. With over 30 years experience, he has a strong R&D background in storage technologies, and is the recipient of many awards in the field. Dr. Tanaka's achievements include the development of the world’s first PMR HDD product, the creation of a spintronics-based Giant Magneto-Resistive head, and work on NAND-based hybrid drives. He plays a research advisory role in Japan’s National Project for spintronics, and is active in many international academic conferences. He holds a PhD and MSEE in electronics engineering from Tohoku University (Japan) and has been a research scientist at the University of Minnesota’s Center for Micromagnetics and Information Technology
Abstract: New 3D Flash Technologies Offer Both Low Cost and Low Power Solutions
New flash applications have a wide variety of requirements, leading to a need for multiple 3D technologies. Wearables, smartphones, automobiles, and the IoT all require low power usage as well as high performance to process data locally. Data center applications require high density and low cost to meet the needs of such data-intensive tasks as real-time analytics, computational genomics, cloud computing, and video and image processing. New technologies such as QLC (Quadruple Level Cell) BiCS FLASH™ offer high density, low-cost solutions, while TSV (Through Silicon Via) NAND offers high performance with significant power reduction. A variety of new approaches are accelerating the data center revolution while also enabling broader compute and storage architectures.
Toshiba, a Fortune Global 500 company, focuses its world-class capabilities in advanced electronic and electrical products and systems into three business areas:
Founded in Tokyo in 1875, today’s Toshiba is at the heart of a global network of 550 consolidated companies employing 188,000 people worldwide, with annual sales surpassing 5.7 trillion yen (US$50 billion). To learn more about Toshiba, visit www.toshiba.co.jp/index.htm More information is available at Toshiba.
Executive Vice President, Memory Technology
Dr. Siva Sivaram is the executive VP of memory technology for Western Digital. He is responsible for the ongoing development of the company's industry-leading NAND flash memory, as well as next-generation technologies, including 3D NAND (BiCS) and 3DReRAM. Sivaram has over 30 years of experience in semiconductors, 3-D memory architectures, process technology, equipment and materials. He has held executive positions at Intel and Matrix Semiconductor, and at SanDisk after its acquisition of Matrix.
Under his leadership, SanDisk created the world’s first cross point 3D memory technology, the world leading 15nm 2D NAND technology, and the industry’s first 256 Gigabit (Gb) 3-bit-per-cell (X3) 48-layer 3D NAND chip.
Sivaram has also been a board member of several start-up firms, entrepreneur-in-residence at Crosslink Capital and XSeed Capital, a research scholar at Matsushita Electric, and an adjunct faculty member at San Jose State University.
Sivaram has published many technical papers as well as a textbook on Chemical Vapor Deposition, and holds several patents. He earned his doctorate and master’s degrees in materials science from Rensselaer Polytechnic Institute (RPI), where he now serves on the Board of Trustees.
Abstract: Creating Storage Class Memory: Learning from 3D NAND
The next great challenge for storage companies will be developing Storage Class Memory (SCM), that is, nonvolatile memory with access rates and other characteristics more like DRAM than flash. However, the enormous promise of SCM in computing environments cannot be fulfilled unless the technology can escape the challenges of scaling. How can such a potentially fast, powerful, and high capacity technology be developed in a way that is both practical and economical enough for widespread commercialization? One approach is to draw on the lessons learned in developing 3D NAND.
The technologies differ, but memory engineers will again need to overcome lithographic complexity, limits in charge storage, and proximity effects of adjacent memory cells. These were all significant hurdles which were overcome during 3D NAND development. So creating SCM should involve equal attention to scaling and cost, and a focus on vertical rather than horizontal layers. Memory engineers must follow such approaches if they are to unleash the enormous promise of SCM.
About Western Digital:
Western Digital is an industry-leading provider of storage technologies and solutions that enable people to create, leverage, experience, and preserve data. The company addresses ever-changing market needs by providing a full portfolio of compelling, high-quality storage solutions with customer-focused innovation, high efficiency, flexibility, and speed. Our products are marketed under the HGST, SanDisk, and WD brands to OEMs, distributors, resellers, cloud infrastructure providers, and consumers. More information is available at Western Digital.
Chief Research Officer, Storage
Balint Fleischer is currently Chief Research Officer at Huawei’s Central Research Institute, where he is responsible for research and technology planning in the storage area. Widely recognized as one of the leading computer architects in the technology industry, he was most recently CTO at startup Parallel Machines, where he developed new architectures for advancing predictive analytics and machine learning. He was previously General Manager and Director of Architectures for Workstation, Storage, and Server Platforms at Intel, where he managed Atom and Xeon server architecture development, including efforts related to 3DXPoint ™ and Rack Scale Architecture. He also had long experience at Sun Microsystems including being VP/CTO Networked Storage Division, where he led the design of next generation storage systems and storage virtualization platforms. He led Sun’s architecture development for many successful low end and midrange server products and was responsible for the company’s InfiniBand effort focusing on enterpriseclustering, I/O, and storage.
He holds several patents, has published articles in many trade and technical journals, is a frequent participant at major conferences and in standards groups, and is often quoted in the trade and technical press.
Abstract: Using Storage Class Memory to Create Scalable Cognitive Computers
Learning systems such as cognitive computers can vastly improve enterprise management and decision making. They are designed to provide new insight derived from analyzing multiple large, diverse data sets.
They use advanced analytics to sense, learn, infer, and interact with the user. They can even participate in joint discovery processes and scenario planning, activities that are far beyond the capabilities of today’s systems.
However, such platforms are very complex and have huge computational requirements. Implementing them requires a new way to process, manage, and share information in which storage class memory (non-volatile storage accessible at memory speeds) plays a key role.
Huawei is a leading global provider of commercial telecom networks, and it is currently serving 45 of the world's top 50 telecom operators. Through continuous customer-centric innovation, Huawei responds quickly to customers' needs with a comprehensive, customized set of offerings. Huawei's products and solutions are deployed in over 140 countries and are supporting the communication needs of one-third of the world's population More information is available at Huawei.
Dr. Zining Wu
Chief Technology Officer
Dr. Wu serves as Chief Technology Officer of Marvell Technology Group Ltd., a position he has held since January 2014. In this role, Dr. Wu is responsible for overseeing all technical aspects of the company including establishing the company’s technical vision and strategic innovation initiatives, and directing R&D project execution and future growth.
Prior to his current position, Dr. Wu was Vice President, Data Storage Technology at Marvell. In this role, he was responsible for leading the engineering group that delivers innovative storage technologies for the hard disk drive (HDD) and solid-state drive (SSD) electronics markets. During his leadership, Marvell announced the world’s first low-density parity-check (LDPC) code based SSD controller that more than triples the reliability of NAND flash; the first native PCIe SSD controller for modular scalability, raising the bar in data storage by delivering an unprecedented level of I/O performance; and the world’s first NVRAM-powered PCIe SSD cache solution providing industry-leading performance for cloud computing data centers. In addition, also during his tenure, Marvell achieved a significant milestone with Western Digital, with the two companies shipping more than one billion hard disk drives with Marvell controllers. Previous to this role, Dr. Wu held various engineering and managerial roles within Marvell, including Vice President of Engineering for wireless communication SoCs, where his team delivered the industry’s best-in-class 802.11ac chips. He joined Marvell in July 1999.
Dr. Wu holds a Bachelor of Science degree in Electronic Engineering from Tsinghua University in Beijing, China, and a Master of Science degree and Ph.D. in Electrical Engineering from Stanford University. Dr. Wu holds over 280 U.S. patents and has published eight technical papers and a book related to data storage technology titled, “Coding and Iterative Detection for Magnetic Recording Channels.”
Abstract: Managing Multi-Tiered Non-Volatile Memory Systems for Cost and Performance
Manufacturers have recently introduced a wide range of new or greatly improved non-volatile memories, including 3D XPoint™, MRAM, RRAM, and specially tuned NAND flash for cold storage. They all have useful characteristics in many applications, and designers will want to take advantage of several or even all of them. The problem is how to scale out the resulting storage and manage data transfer between different types (or tiers) to achieve the highest performance at the lowest cost. Hardware-based cache management engines can provide high-performance solutions for the higher-performing tiers. For the lower tiers, designers can reduce overall system cost by concatenating expandable controllers to manage larger drives rather than developing new ASICs or looking for new standard products.
Marvell is a global leader in providing complete silicon solutions. From storage to cloud infrastructure, Internet of Things (IoT), connectivity and multimedia, Marvell’s diverse product portfolio aligns complete platform designs with industry-leading performance, security, reliability and efficiency. At the core of the world’s most powerful consumer, network and enterprise systems, Marvell empowers partners and their customers to always stand at the forefront of innovation, performance and mass appeal. By providing people around the world with mobility and ease of access to services adding value to their social, personal and work lives, Marvell is committed to enhancing the human experience. More information is available at Marvell.
Sr VP, Head of NAND Division
Kyo Won Jin is Senior Vice President and Head of the NAND Division at SK Hynix. He is responsible for all NAND development activities from product planning through development and productization. He previously oversaw the DRAM Product Development Group and NAND Product Planning Office.
With his extensive 30 years of experience in DRAM and NAND design, product planning, and productization, Kyo Won has played a major role in SK Hynix’s turnaround, helping the company become a global leader. He also served as the Head of the Semiconductor Planning Office at SK Telecom (a wireless telecommunications operator in Korea) for two years where he acquired broad knowledge of business development in the telecom industry.
In recognition of his contributions to the semiconductor industry, he was awarded the Presidential Citation of the Republic of Korea. He holds a BS in physics from Seoul National University.
Abstract: NAND Satisfies Both Performance-Based and Cost-Sensitive Enterprise Applications
NAND flash satisfies both performance and capacity needs in today’s enterprise market. On the one hand, NAND flash can provide the low latency and high bandwidth required by high-performance computing, analytics, security, database, video, media and entertainment, and financial applications. On the other hand, it can offer low-enough costs, particularly when operating expenses are considered, to replace hard drives in ordinary business applications and even in archiving, disaster recovery, and backup. 3D NAND can extend NAND’s capabilities even further in both areas, and means that flash will also be the major storage mechanism in emerging areas such as clouds and hyperconverged infrastructures.
About SK Hynix:
SK Hynix, headquartered in Korea, is a top tier semiconductor supplier providing Dynamic Random Access Memory chips (“DRAM”), flash memory chips ("NAND Flash") and CMOS image sensors to a wide range of global customers. More information is available at SK Hynix.
Director, NAND Product Marketing
Executive VP R&D, Samsung Memory Business
Jaeheon Jeong is Executive Vice President of the Samsung Memory Business and currently oversees solution product R&D, which delivers flash storage products ranging from micro-SD cards to enterprise SSDs. Before taking on his current position, he was with Samsung’s controller team and later led its software development team. He initiated major industry-leading projects such as those involving the industry-first AHCI PCIe SSD, enterprise NVMe SSDs, and UFS solutions. Recently, he led the push behind the large-capacity storage that best utilizes V-NAND technology, including the world largest 2.5” 16TB SAS SSD. He also has played a key role in expediting the HDD to SSD transition.
Dr. Jeong has 25 years experience in the computer and semiconductor industry. He began as a hardware engineer in the Samsung Computer Division. He later worked for IBM and Intel in server computer architecture and performance before rejoining Samsung in 2010. He holds a PhD. in Electrical Engineering and an MS in Computer Engineering from the University of Southern California, and a BSEE from Korea University.
Abstract: 3D NAND is the Leadership Technology for Server Storage
To meet the demands of big data, cloud computing and real-time analysis, data center managers, enterprise customers and ecosystem partners require non-volatile memory with increased performance, higher densities, smaller footprints and a strong TCO. 3D NAND is the production methodology best able to meet those marketplace needs. Advances already in the pipeline will provide higher capacity devices with better operating characteristics for server workloads and for the ever-increasing needs of clouds and mega-websites. In particular, such devices will be especially well-suited for rapidly emerging NVMe SSD technology. Special requirements for low power consumption, small size, low latency, and ultra-high capacity will also be met. Technological progress will continue unabated along all of these avenues as 3D NAND reaches its full potential.
Samsung Semiconductor is a wholly owned subsidiary of Samsung Electronics. As the leader in advanced memory solutions for home, mobile, and office applications, Samsung Semiconductor is the world’s largest producer of DRAM, flash, SRAM, and high-end graphics memory. In addition, it is a major global producer of optical disk drives, hard drives, system logic, sensors, and controllers. Samsung also has emerged as the largest producer of LCD displays, providing the widest range of products for TVs, desktop monitors, notebook/netbook PCs, and mobile devices. Furthermore, Samsung provides fabless and fab-lite customers with state-of-the-art design and manufacturing at 90, 65, 45, and 32/28 nanometers. More information is available at Samsung.
Vice President & Business Unit Manager, Performance Storage
Derek Dicker is VP/Business Unit Manager for the Performance Storage Business Unit (PSBU) at Microsemi. He has P&L responsibility for PCIe storage switches, SSD controllers, and NV-RAM drives. Dicker previously was VP Performance Solutions Group at PMC-Sierra before its acquisition by Microsemi. At PMC, he led the team that introduced the first and fastest merchant enterprise NVMe PCIe SSD controller and NV-RAM drive portfolio, and introduced a new category of PCIe storage switches, including the first NVMe JBOD-optimized switch solutions.
Prior to launching the performance storage business unit, Dicker headed marketing for PMC’s enterprise storage division, where he and his team defined, launched, and ramped the company’s 12G SAS product portfolio, featuring industry-leading port count and performance SAS/SATA RoCs/IoCs and expanders. Before joining PMC, he was vice president and general manager for the networking division at IDT. He spent the early part of his career at Intel in sales and marketing positions, including the role of chief of staff/technical assistant to the executive vice president and general manager of the Intel Communications Group. Dicker earned a BS in Computer Science and Engineering from UCLA, and later attended the Stanford Executive Program.
Abstract: The NVMe Inflection Point
NVMe SSDs are currently enabling a new level of performance for primary storage in enterprise and hyperscale applications, while maintaining full conformity with a widely used standard interface. Competition and technology innovation have enabled cost-effective products to come to market featuring varied form factors, capacities, and capabilities. Meanwhile, storage vendors are delivering enclosures optimized for aggregating and managing large numbers of NVMe SSDs. And lastly, the industry is rallying around standards efforts to help achieve interoperability and volume economics. Such initiatives include fabrics, drive- , and enclosure-management efforts. Opportunities still exist for more innovation in this exciting new tier of storage, involving both hardware and software.
Microsemi Corporation offers a comprehensive portfolio of semiconductor and system solutions for aerospace & defense, communications, data center and industrial markets. Products include high-performance and radiation-hardened analog mixed-signal integrated circuits, FPGAs, SoCs and ASICs; power management products; timing and synchronization devices and precise time solutions, setting the world’s standard for time; voice processing devices; RF solutions; discrete components; enterprise storage and communication solutions, security technologies and scalable anti-tamper products; Ethernet solutions; Power-over-Ethernet ICs and midspans; as well as custom design capabilities and services. Microsemi is headquartered in Aliso Viejo, Calif., and has approximately 4,800 employees globally. More information is available at Microsemi.
President, Cloud Systems and Silicon Group
Phil Brace is president of Seagate’s Cloud Systems and Silicon Group, responsible for the full breadth of enterprise storage systems, flash technology, and silicon R&D. This includes Seagate's growth initiatives in cloud computing and solid state technology, optimized to build storage solutions for next-generation workloads and new levels of performance for OEM and cloud customers. Brace brings over 20 years of semiconductor and systems experience to his current role.
Brace previously headed up Seagate’s Electronics Solutions organization. Before joining Seagate, he spent nine years at LSI in senior technology and business leadership roles, including executive vice president of the Storage Solutions Group, senior vice president and general manager of the Storage Peripherals Division, and senior vice president of the marketing communications, corporate communications, investor relations and strategic planning functions.
Earlier in his career, Brace worked at Intel, where he served as general manager of Server Platforms Group Marketing, leading the organization charged with worldwide product and technical marketing, including demand creation for technologies across all of Intel’s server business.
Brace received his MSEE from California State University, Sacramento and his Bachelor’s degree in Computer Engineering from the University of Waterloo (Canada).
Abstract: Design Efficient Storage Systems with Both Flash and HDDs
Efficient storage systems today involve both flash and HDDs. Flash provides fast access but limited capacity. HDDs provide huge capacities with relatively long access times. Proven, effective caching software will allow the best use of both kinds of storage. Direct transfers to and from storage elements (without processor involvement) will also speed up systems generally. The latest SMR HDDs can provide cost-effective storage in situations such as archiving that can tolerate longer access times. Designers must combine the latest flash and HDD technologies to produce optimized storage systems with the best cost per TB and per IOP. The end result will be better ROI for such applications as real-time data analysis, big data processing, and hyperscale data centers.
Seagate is the global leader in data storage solutions, developing amazing products that enable people and businesses around the world to create, share and preserve their most critical memories and business data. Over the years the amount of information stored has grown from megabytes all the way to geopbytes, confirming the need to successfully store and access huge amounts of data. As demand for storage technology grows the need for greater efficiency and more advanced capabilities continues to evolve..
Today data storage is more than just archiving; it’s about providing ways to analyze information, understand patterns and behavior, to re-live experiences and memories. It’s about harnessing stored information for growth and innovation. Seagate is building on its heritage of storage leadership to solve the challenge of getting more out of the living information that’s produced everyday. What began with one storage innovation has morphed into many systems and solutions becoming faster, more reliable and expansive. No longer is it just about storing information; it is about accessing and interpreting information quickly, accurately and securely. More information is available at Seagate.
Daniel Cobb, Sr.
EMC Fellow, Vice President – New Media Strategy
Danny Cobb is a Fellow and VP New Media Strategy at EMC, where he leads global technology strategy and architecture for emerging storage media. His work is central to EMC’s leadership in the IT industry’s transition from hard drives to flash memory for primary storage. Cobb is currently responsible for the technological readiness, investment priorities, and product strategies that prepare EMC’s customers for the coming wave of emerging memories.
Danny also has led early stage product and technology initiatives as CTO of the Flash Technologies Division, Converged Infrastructure, and Data De-Duplication. His special interest areas include server and storage design, system performance, and the hardware/software interface.
Cobb was previously a founding team member of Storigen Systems (acquired by EMC), where he led system architecture and platform design for an innovative rich media delivery engine. His experience includes working on file systems at Avid Technology and operating systems at Digital Equipment.
Cobb earned a BS in Computer Science from the University of Vermont. He holds several patents in distributed systems and storage design, and he often advises early stage companies and venture capital firms.
Abstract: Flash Storage Meets Persistent Memory: The Modern Data Center Changes Forever!
The backdrop of big data volume, cloud data ubiquity, and IoT data velocity is putting tremendous pressure on data centers to increase their performance and scalability rapidly. The key to doing this is persistent memory, storage that operates at memory speeds. It will allow server designers and storage architects to define a new memory-centric architecture (MCA). The result is to exploit breakthrough technology capabilities to create a next-generation data center with real-time, high-frequency, low-latency data management frameworks.
EMC Corporation is a global leader in enabling businesses and service providers to transform their operations and deliver IT as a service. Fundamental to this transformation is cloud computing. Through innovative products and services, EMC accelerates the journey to cloud computing, helping IT departments to store, manage, protect and analyze their most valuable asset – information – in a more agile, trusted and cost-efficient way. More information is available at EMC.
Thomas Isakovich is CEO and founder of Nimbus Data, overseeing product strategy and development, sales, marketing, and manufacturing operations. Thomas founded Nimbus Data to build the next great storage systems company by combining smart software with non-volatile memory technology, offering superior data delivery, protection, and operating efficiency. Prior to Nimbus Data, Thomas was CEO and founder of TrueSAN Networks, where he raised over $34 million and led the development of storage virtualization and multi-vendor SAN management software. At Oracle spin-off Network Computer (also known as Liberate Technologies), Thomas oversaw product marketing for the company’s line of thin-client workstations and server software. At IBM’s Almaden Research Center, Thomas developed new technologies for simplifying and enhancing human-computer interaction. A frequent speaker on technology and entrepreneurship at universities and industry events, Thomas earned a B.A. with Honors in political science from Stanford University.
Abstract: Scaling Flash Storage Efficiently in the Exabyte Era
The amount of data organizations have is increasing at an incredible rate. Petabytes are commonplace, and exabytes are the next frontier. Flash memory will be essential for processing huge amounts of data at reasonable speeds, while at the same time recognizing constraints on cost, power, and space. But how do we scale all that flash? Simple scale-out methods increase latency and energy consumption. However, scale-up arrays are expensive, and scale-out clusters are difficult to manage. A federated approach is the best answer. It allows for significant expansion without raising latency or energy consumption greatly. The idea is to do as much processing as possible at the lowest level possible, minimizing transfers between units and thus keeping costs, latency, and energy consumption under control.
About Nimbus Data:
Nimbus Data develops an award-winning portfolio of all-flash memory storage systems purpose-built for virtualization, databases, cloud infrastructure, and technical computing. Our solutions, engineered from the ground up for flash memory, combine patent-pending hardware innovation and comprehensive data management software to deliver the industry’s best performance, energy efficiency, scalability, and total cost of ownership. More information is available at Nimbus Data.
Dr. George Minassian is co-founder and CEO of Crossbar, where he leads the development of low-power, high-performance, high-density RRAM memory technology. Dr. Minassian has a proven track record of developing commercially successful, leading-edge memory and communications products. Before joining Crossbar, he worked at Spansion, most recently as VP System and Software Engineering. While at Spansion, he led the $1.2 billion flash memory business targeting the cellular wireless market. He also developed new industry standards in emerging technology and established a successful track record for large-scale product development and management. He was previously Director of Wireless Engineering at Advanced Micro Devices, where he developed the industry’s first CMOS RF process and complete 802.11b/a chipset and reference designs. Dr. Minassian holds a Ph.D. in Electrical Engineering and Computer Science from the University of Texas at Austin.
Abstract: RRAM: Your Key to Creating Faster, More Scalable Storage
Imagine a high capacity memory technology with 100X lower latency, one-twentieth the power consumption, and 1000X higher endurance than today’s NAND flash. Resistive RAM (RRAM) is that technology. Today’s devices are available now to serve the low energy internet-of-things applications, secure or general-purpose micro controllers, sensors and actuators connected systems that require extremely low power, high performance, endurance and reliability. Devices on the horizon will be suitable for high-density, low latency storage class memory for data centers and media-rich mobile applications. RRAM also has a path to 5 nm solutions where NAND flash will have difficulty competing. Now is the time for storage system designers to consider RRAM for their next-generation designs
Crossbar is the leader in RRAM technology, widely accepted as the front-runner to replace traditional flash technology in future storage systems. Delivering terabyte storage on a postage stamp-sized chip, with power low enough for massive adoption throughout the Internet of Things, Crossbar RRAM is easy to tailor for a broad range of applications. From embedded memory on SOCs for wearables, to very high density SSDs for cloud data centers, Crossbar is ushering in a new era of storage innovation More information is available at Crossbar.
Paul Prince is CTO at Mangstor, where he provides strategic product and business planning, technology and product development, talent development and retention, and strategic communications with company top executives, press, analysts, and customers. A successful technology professional with over 20 year experience leading architecture and technology development at premier computer technology and product companies, he has demonstrated deep technology insight, proven leadership in functional and strategic initiative management roles, and the ability to translate leading edge technology trends into real-world products.
Before joining Mangstor, he was CTO at Dell, where he was responsible for technology direction, investigation, and development for Dell servers. Prior to working at Dell, Paul was Platform Architecture Director at Intel, where he was executive leader of all server system architecture and technology and led a geographically diverse team of over 100 engineers and architects. He holds an MSEE from the University of Utah
Abstract: NVMe over Fabrics Offers Top Performance for Real Time Analytics
The demand for real-time analysis of huge datasets provides a major challenge for storage providers. Flash-based storage must offer more bandwidth, higher density, and lower prices. One solution is to offer NVMe/PCIe SSDs connected via high-speed fabrics. The result is a powerful combination of scalability, fully shared resources, high bandwidth, and exceptionally low latency, as compared to alternatives such as SAN-connected flash arrays. Products are already available that are optimized for high performance computing and big data problems. Such storage array products can deliver data to analytics algorithms at the bandwidth required for real-time operation.
Mangstor provides industry leading storage products based on the industry standard NonVolatile Memory Express (NVMe). The MX6300 PCIe card SSD is the highest performance and lowest latency SSD in the industry, for use in servers and storage products. The NX6320 Flash array is the industry’s first available product supporting the NVMe over Fabric interface, with support for Ethernet and InfiniBand fabrics. NVMe over Fabrics offers considerable performance and latency advantages over iSCSI and Fiber Channel Flash storage solutions. More information is available at Mangstor.