Showing posts with label Silicon. Show all posts
Showing posts with label Silicon. Show all posts

Wednesday, October 31, 2018

Netronome proposes open "chiplets" for domain specific workloads

Netronome unveiled its open architecture for domain-specific accelerators .

Netronome is collaborating with six leading silicon companies, Achronix, GLOBALFOUNDRIES, Kandou, NXP, Sarcina and SiFive, to develop this open architecture and related specifications for developing chiplets that promise to reduce silicon development and manufacturing costs.

The idea is fo chiplet-based silicon to be composed using best-of-breed components such as processors, accelerators, and memory and I/O peripherals using optimal process nodes. The open architecture will provide a complete stack of components (known good die, packaging, interconnect network, software integration stack) that lowers the hardware and software costs of developing and deploying domain-specific accelerator solutions. Implementing open specifications contributed by participating companies, any vendor’s silicon die can become a building block that can be utilized in a chiplet-based SoC design.

“The end of Moore’s Law will increase the use of domain-specific accelerators to meet power-performance requirements in cloud infrastructure, network infrastructure and IoT/wireless edge applications,” said Bob Wheeler, principal analyst, The Linley Group. “With its modular approach, the open domain-specific accelerator architecture could change the chiplet paradigm from single-vendor solutions to a world of choice, thereby enabling OEMs and operators to develop and deploy advanced SoC solutions.”

“Netronome’s domain-specific architecture as used in its Network Flow Processor (NFP) products has been designed from the ground up keeping modularity, and economies of silicon development and manufacturing costs as top of mind,” said Niel Viljoen, founder and CEO at Netronome. “We are extremely excited to collaborate with industry leaders and contribute significant intellectual property and related open specifications derived from the proven NFP products and apply that effectively to the open and composable chiplet-based architecture being developed in the ODSA Workgroup.”

https://www.netronome.com/

Sunday, October 28, 2018

Broadcom intros first full-duplex G.fast

Broadcom introduced the first full-duplex modem implementation based on G.fast.

Broadcom said the innovation builds on the advantages of G.fast for high-density residential environments. Full-duplex utilizes the entire frequency spectrum of G.fast simultaneously in both the downstream and upstream communication paths, effectively doubling the simplex capacity of G.fast. Depending on the nature of an operator’s copper plant, this will result in either a doubling of the peak bandwidth, or the maximum distance from the distribution point to the customer premise, as compared with the existing G.fast standard.

Broadcom’s full-duplex implementation is available today on the production-released BCM65400 device family. 

Broadcom is working with the next-generation G.mgfast initiative within the ITU to bring a fully-standardized full-duplex solution to the broadband market, pending completion and ratification of the new standard.

“We are excited to pioneer another innovative copper solution, one that effectively doubles the performance of current G.fast,” said Greg Fischer, senior vice president and general manager, Broadband Carrier Access, Broadcom. “Full-duplex technology will provide telecom operators with yet another option to leverage the investment in their existing copper networks, with a competitive, minimal capex multi-gigabit broadband service for many years to come.”

Monday, October 15, 2018

Samsung intros automotive app processors and image sensor

Samsung Electronics introduced two new automotive solution brands -- Samsung Exynos Auto and Samsung ISOCELL Auto – targeting application processors and image sensors.

“Samsung's new automotive brand solutions, Exynos Auto and ISOCELL Auto, bring Samsung's market-proven technologies to automotive applications with enhanced features and durability required by the market,” said Kenny Han, vice president of Samsung’s Device Solutions Division. “With fast telecommunication, accurate sensing and powerful processing capabilities, Samsung’s Auto-branded solutions will enable new driving experiences to next-generation smart vehicles.”

Samsung’s Exynos automotive processors bring the efficient qualities of Exynos’ mobile versions and are designed to meet industry requirements for harsher environments for automotive infotainment, advanced driving assistance systems (ADAS) and telematics.

Samsung’s ISOCELL image sensors feature the company’s innovative pixel isolation technology. The automotive grade ISOCELL sensors provide greater visibility of the road and surroundings even in low-light environments while enabling more precise identification of objects.

Monday, October 1, 2018

Intel touts new Xeon server performance benchmarks

Intel reported 95 new performance world records for its Xeon Scalable processors as achieved in servers from major original equipment manufacturers, ranging from single-socket systems up to eight-socket systems.

The new performance benchmarks were achieved in a variety of server platforms, ranging from general computing workloads running on single-socket systems to advanced technical computing and big data analytics workloads running on eight-socket systems. All systems tested include mitigations for Spectre and Meltdown.

“I’m extremely proud of the 95 world record performance benchmarks that our partners have delivered, but even more delighted to see the real-world performance that our customers are achieving on the fastest ramping Intel Xeon processor family in history,”
stated Lisa Spelman, vice president and general manager of Intel Xeon products and data center marketing.

Last week, Intel CFO and Interim CEO, Bob Swan, disclosed that its data-centric businesses grew 25 percent through June, and cloud revenue grew a whopping 43 percent in the first six months of the year. Responding to increasing demand for server and PC processors, Intel updated its CAPEX guidance for the year, saying it now plans to invest a record $15 billion in capital expenditures in 2018, up approximately $1 billion from the beginning of the year. Intel is investing $1 billion into its 14nm manufacturing sites in Oregon, Arizona, Ireland and Israel. Swan also cited progress with 10nm, stating that volume production will begin in 2019.

https://www.intel.com/content/www/us/en/benchmarks/server/xeon-scalable/1st-gen-xeon-world-record.html

Monday, September 10, 2018

Intel acquires NetSpeed Systems for interconnect fabric expertise

Intel has acquired NetSpeed Systems, a start-up based in San Jose, California, for its system-on-chip (SoC) design tools and interconnect fabric intellectual property (IP). Financial terms were not disclosed.

Intel said NetSpeed’s highly configurable and synthesizable offerings will help it more quickly and cost-effectively design, develop and test new SoCs with an ever-increasing set of IP.

NetSpeed provides scalable, coherent, network-on-chip (NoC) IP to SoC designers. NetSpeed’s NoC tool automates SoC front-end design and generates programmable, synthesizable high-performance and efficient interconnect fabrics. The company was founded in 2011.

The NetSpeed team is joining Intel’s Silicon Engineering Group (SEG) led by Jim Keller. NetSpeed co-founder and CEO, Sundari Mitra, will continue to lead her team as an Intel vice president reporting to Keller.

“Intel is designing more products with more specialized features than ever before, which is incredibly exciting for Intel architects and for our customers. The challenge is synthesizing a broader set of IP blocks for optimal performance while reining in design time and cost. NetSpeed’s proven network-on-chip technology addresses this challenge, and we’re excited to now have their IP and expertise in-house,” stated Jim Keller, senior vice president and general manager of the Silicon Engineering Group at Intel.

Wednesday, September 5, 2018

MACOM samples single-chip, 100G bidirectional optical device

MACOM Technology Solutions introduced an integrated, single-chip transmit and receive solution for short reach 100G optical transceivers, active optical cables (AOCs) and onboard optical engines.

The new MALD-37845 device, which is designed for use in CPRI, 100G Ethernet, 32G Fibre Channel and 100G EDR InfiniBand applications, integrates seamless integration of four-channel Transmit and Receive Clock Data Recovery (CDRs), four Transimpedance Amplifiers (TIAs) and four Vertical-Cavity Surface-Emitting Laser (VSCEL) drivers.

It supports a full range of data rates from 24.3 to 28.1 Gbps, it supports interoperability with a variety of VCSEL lasers and photodetectors, and is firmware-compatible with earlier-generation MACOM solutions.

Sampling is underway with production availability planned for the first half of 2019.

MACOM said the device will provide customers with a low power, single-chip solution ideally suited for small form-factor optical subassemblies.

“Optical module and AOC providers are under tremendous pressure to enable their customers with 100G connectivity at volume scale,” said Marek Tlalka, Senior Director of Marketing, High-Performance Analog, MACOM. “We believe that the MALD-37845 overcomes the integration and cost challenges inherent to legacy multi-chip offerings, providing an elegant, high-performance solution for short reach 100G applications.”

http://www.macom.com

Monday, September 3, 2018

NXP acquires OmniPHY for automotive Ethernet

NXP Semiconductors has acquired OmniPHY, a provider of automotive Ethernet subsystem technology. Financial terms were not disclosed.

NXP said OmniPHY's interface IP and communication technology along with NXP’s own automotive portfolio will form a “one-stop shop” for automotive Ethernet. The companies’ technology synergies will center on 1.25-28Gbps PHY designs and 10-, 100- and 1000BASE-T1 Ethernet in advanced processes.

“Our heritage in vehicle networks is rich and with our leadership positions in CAN, LIN, and FlexRay, we hold a unique viewpoint on automotive networks,” said Alexander E. Tan, vice president and general manager of Automotive Ethernet Solutions, NXP. “The team and technology from OmniPHY give us the missing piece in an extensive high-bandwidth networking portfolio.”

"We are very excited to join NXP – a leader in automotive electronics, for a front-row seat to the autonomous driving revolution, one that will deliver profound change to the way people live,” said Ritesh Saraf, CEO of OmniPHY. “The combination of our teams and technology will accelerate and advance the delivery of automotive Ethernet solutions providing our customers with high quality and world-class automotive Ethernet innovation."

Tuesday, August 28, 2018

Lattice Semiconductor appoints AMD exec as its new CEO

Lattice Semiconductor appointed Jim Anderson as its new President and Chief Executive Officer, and to the company’s Board of Directors. He most recently served as at Advanced Micro Devices (AMD) as the General Manager and Senior Vice President of the Computing and Graphics Business Group.

Jeff Richardson, Chairman of the Board, said, “On behalf of the Board, we are pleased to announce the appointment of Jim Anderson as Lattice’s new President and Chief Executive Officer. Jim brings a strong combination of business and technical leadership with a deep understanding of our target end markets and customers. The transformation he drove of AMD’s Computing and Graphics business over the past few years is just a recent example of his long track record of creating significant shareholder value.

President Trump blocks sale of Lattice Semi citing National Security

President Trump signed an order blocking the sale of Lattice Semiconductor to Canyon Bridge Capital Partners on national security grounds. The issue was referred to the President by the Committee on Foreign Investment in the United States (CFIUS) due to concerns regarding China Venture Capital Fund Corporation Limited and its interest in Canyon Bridge Capital Partners.

Darin G. Billerbeck, CEO of Lattice Semiconductor, issued the following statement:

“The transaction with Canyon Bridge was in the best interests of our shareholders, our customers, our employees and the United States. We also believe our CFIUS mitigation proposal was the single most comprehensive mitigation proposal ever proposed for a foreign transaction in the semiconductor industry and would have maximized United States national security protection while still enabling Lattice to accept Canyon Bridge’s investment and double American jobs. While it is disappointing that we were not able to prevail, the Board and I would like to thank Canyon Bridge for their support during this time.”

https://www.whitehouse.gov/the-press-office/2017/09/13/order-regarding-proposed-acquisition-lattice-semiconductor-corporation

Private Equity Firm Acquires Lattice Semi for $1.3 Billion - FPGAs

Canyon Bridge Capital Partners agreed to acquire all outstanding shares of Lattice Semiconductor Corporation (NASDAQ:LSCC) for approximately $1.3 billion inclusive of Lattice’s net debt, or $8.30 per share in cash. This represents a 30% premium to Lattice’s last trade price on November 2, 2016, the last trading day prior to announcement.

Lattice supplies low power FPGA, video ASSP, 60 GHz millimeter wave, and IP products to the consumer, communications, industrial, computing, and automotive markets worldwide. The company is based in Portland, Oregon.

Monday, August 27, 2018

Global Foundries hits the pause button on 7nm FiFET

GLOBALFOUNDRIES announced a strategic shift away from 7nm process technology and toward "clients in high-growth markets."

Specifically, GF's 7nm FinFET program is on hold indefinitely and the company will shift development resources to make its 14/12nm FinFET platform more relevant, with a focus on new features such RF, embedded memory, and low power.

The company announced a workforce reduction in conjunction with this move, although a number of top technologists will be redeployed on 14/12nm FinFET derivatives and other differentiated offerings.

GF also announced plans to establish its ASIC business as a wholly-owned subsidiary, independent from the foundry business.

Tom Caulfield, GF's CEO, stated “Demand for semiconductors has never been higher, and clients are asking us to play an ever-increasing role in enabling tomorrow’s technology innovations. The vast majority of today’s fabless customers are looking to get more value out of each technology generation to leverage the substantial investments required to design into each technology node. Essentially, these nodes are transitioning to design platforms serving multiple waves of applications, giving each node greater longevity. This industry dynamic has resulted in fewer fabless clients designing into the outer limits of Moore’s Law. We are shifting our resources and focus by doubling down on our investments in differentiated technologies across our entire portfolio that are most relevant to our clients in growing market segments.”

https://www.globalfoundries.com

Wednesday, August 22, 2018

Samsung debuts NB-IoT processor/modem/GNSS

Samsung Electronics Co. unveiled its new narrowband (NB) Internet-of-Things (IoT) chip.  The Exynos i S111 integrates a modem, processor, memory and Global Navigation Satellite System (GNSS) into a single chip design to enhance efficiency and flexibility for connected device manufacturers.

Samsung said its NB-IoT solution offers extremely wide coverage, low-power operation, accurate location feedback and strong security. It is optimized for real-time tracking applications such as safety wearables or smart meters.

The Exynos i S111 incorporates a modem capable of LTE Rel. 14 support that can transmit data at 127-kilobits-per-second (kbps) for downlink and 158kbps uplink, and can operate in standalone, in-band and guard-band deployments.

NB-IoT adopts a data retransmission mechanism that continuously transmits data until a successful transfer, or up to a set number of retransmits. With a high number of these retransmit sessions, the S111 is able to cover the distance of 10-kilometers (km) or more.

“IoT will be able to evolve to offer new features beyond the conventional household space with IoT-dedicated solutions that present a broad range of opportunities,” said Ben Hur, vice president of System LSI marketing at Samsung Electronics. “Exynos i S111’s highly secure and efficient communication capabilities will bring more exciting NB-IoT applications to life.”

http://www.samsung.com/exynos

Sunday, July 29, 2018

Caixin: The Wake-Up Call for China’s Chip Industry

The near-death experience of ZTE due to the 88-day ban imposed the U.S. Department of Commerce is a wake-up call for China's semiconductor industry, according to an article in the business journal Caixin.

The article the foreign joint venture model, Chinese investments/acquisitions of tech companies abroad, and government-supported initiatives to build a strong, domestic ecosystem for semiconductor design and fabrication.

https://www.caixinglobal.com/2018-07-29/the-wake-up-call-for-chinas-chip-industry-101309608.html

Wednesday, July 18, 2018

Intel celebrates its 50th birthday

Intel marked the 50th anniversary of its founding on July 18, 1968 by Robert Noyce and Gordon Moore.

A ceremony in New York City and Santa Clara, California was presided over by Bob Swan, Intel’s chief financial officer and interim chief executive officer. Instead of ringing the Nasdaq opening bell, Intel’s well-known five-note jingle was played accompanied by the Intel bong.

“We’re only just beginning to explore the impact that technologies like autonomous driving, artificial intelligence and quantum computing can have. And over the next several decades, we’ll create computing platforms that make the world a better place — from safer and smarter models of mobility to the eradication of disease,” stated Bob Swan, Intel CFO and interim CEO.

Intel also set a new record for most synchronized drones flying in formation -- 500.


Wednesday, July 4, 2018

Baidu develops its own AI chip, rolls out first autonomous bus

At its second annual developer conference in Beijing this week, Baidu unveiled its "Kunlun" processor for AI applications.


Technical details on the new Kunlun silicon were scarce, but the company said its cloud-to-edge AI chip is built to accommodate high-performance requirements of a wide variety of AI scenarios, including deep learning and facial recognition.

Baidu is known to be developing FPGA designs for a number of years.

Baidu also announced volume production of China’s first commercially deployed fully autonomous bus. The first 100 "Apolong" buses are ready for the road.


Thursday, May 31, 2018

Marvell posts revenue of $605 billion, next quarter excludes $7m is sales to ZTE

Marvell Technology Group reported revenue for its first quarter of fiscal 2019 was $605 million, which exceeded the midpoint of the Company's guidance provided on March 8, 2018. GAAP net income from continuing operations for the first quarter of fiscal 2019 was $129 million, or $0.25 per diluted share. Non-GAAP net income from continuing operations for the first quarter of fiscal 2019 was $165 million, or $0.32 per diluted share. Cash flow from operations for the first quarter was $129 million.

"Fiscal 2019 is off to a strong start, driven by the performance of our storage, networking and connectivity businesses which grew 7% year over year in Q1. Marvell's R&D engine is executing well, and our newly announced products are fueling a growing design win pipeline," said Marvell President and CEO Matt Murphy. "Overall, I'm pleased with the results and thank the entire Marvell team for their effort and contribution."

Revenue for the company's second quarter of 2019 is expected to be $600 million to $630 million. The guidance range excludes approximately $7 million in revenue from a Chinese OEM due to the trade restrictions imposed by the U.S. government.

Monday, May 21, 2018

IDC forecasts semiconductor growth of 7.7%, reaching $450B in 2018

IDC is predicting that worldwide semiconductor revenue will grow for the third consecutive year in 2018 to $450 billion, up 7.7% over 2017, down from 24% in 2017.

IDC's new Semiconductor Applications Forecaster (SAF) also forecasts that semiconductor revenues will log a compound annual growth rate (CAGR) of 2.9% from 2017-2022, reaching $482 billion in 2022.

"Market consolidation in the semiconductor industry over the past five years continues to shape the competitive landscape for semiconductor suppliers as each company continues to refine its core markets and make acquisitions to find new and emerging sectors for growth. The pace of change and technology is expected to accelerate as machine learning and autonomous systems enable a more diverse set of architectures to address the opportunity. This will fuel the engine of growth for semiconductor technology over the next decade," said Mario Morales, program vice president, Semiconductors at IDC.

Some key findings from IDC's Semiconductor Application Forecaster (excluding memory) include:


  • Semiconductor revenue for the computing industry segment will decline 4.0% this year and will show a negative CAGR of -0.7% for the 2017-2022 forecast period. Two bright spots for the computing segment are computing and enterprise SSDs, growing in high double digits and 9.8% CAGR respectively for 2017-2022.
  • Semiconductor revenue for the mobile wireless communications segment will grow 5.5% year over year this year with a CAGR of 5.8% for 2017-2022. Semiconductor revenue for 4G mobile phones will experience an annual growth rate of 10.9% in 2018 and a CAGR of 3.1% for 2017-2022. 5G will also drive growth in the later part of the forecast as the technology becomes mainstream by the middle of the next decade.
  • Communications infrastructure semiconductors are forecast to grow at a 1.7% CAGR from 2017-2022 with the strongest growth coming from consumer networks.
  • The automotive market and the industrial markets will continue to be the leading areas of growth for the semiconductor market throughout the forecast period, growing at a 9.6% and 6.8% CAGR from 2017-2022. 

Thursday, May 17, 2018

Tachyum announces its Universal Processor Platform

Tachyum, a start-up based in San Jose, California with offices in Slovakia, unveiled its new processor family – codenamed “Prodigy” – that combines the advantages of CPUs with GP-GPUs, and specialized AI chips in a single universal processor platform. The company says its processor architecture attains ten times the processing power per watt compared to conventional designs.

A key innovation of the design is the ability to connect very fast transistor with very slow wires, but technical details on the device physics have not yet been disclosed.

The universal processor promises programming ease comparable to a CPU with performance and efficiency comparable to GP-GPU. It is designed to handle hyperscale workloads, AI, HPC, and other demanding applications

Tachyum claims its Prodigy universal processor will enable a super-computational system for real-time full capacity human brain neural network simulation by 2020. One target application would be the real-time Human Brain Project, where there’s a need for more than 1019 Flops (10,000,000,000,000,000,000 floating-point operations per second - 10 exaflop).

“Rather than build separate infrastructures for AI, HPC and conventional compute, the Prodigy chip will deliver all within one unified simplified environment, so for example AI or HPC algorithms can run while a machine is otherwise idle or underutilized,” said Tachyum CEO Dr. Radoslav ‘Rado’ Danilak. “Instead of supercomputers with a price tag in the hundreds of millions, Tachyum will make it possible to empower hyperscale datacenters to produce more work in a radically more efficient and powerful format, at a lower cost.”

“Despite efficiency gains from virtualization, cloud computing, and parallelism, there are still critical problems with datacenter resource utilization particularly at a size and scale of hundreds of thousands of servers,” said Christos Kozyrakis, professor of electrical engineering and computer science at Stanford, who leads the university’s Multiscale Architecture & Systems Team (MAST), a research group for cloud computing, energy-efficient hardware, and operating systems. “Tachyum’s breakthrough processor architecture will deliver unprecedented performance and productivity.” Kozyrakis is a corporate advisor to Tachyum.

http://tachyum.com


  • Tachyum is headed by Dr. Radoslav ‘Rado’ Danilak, who previously was founder and CEO of Skyera, a supplier of ultra-dense solid-state storage systems, acquired by WD in 2014. He also was cofounder and CTO of SandForce, which was acquired by LSI in 2011 for $377M. Its cofounders include Rodney Mullendore (previously Sandforce, Nishan Systems, Sandia National Labs); Igor Shevlyakov (previously Skyera); Ken Wagner (previously Wave Computing, Silicon Analystics and Theseus Logic).
  • Tachyum is funded by IPM Growth, the venture capital division of InfraPartners Management LLP.

Monday, April 30, 2018

Semiconductor sales top $111 billion in Q1, up 20%

Worldwide sales of semiconductors reached $111.1 billion during the first quarter of 2018, an increase of 20 percent compared to the first quarter of 2017, but 2.5 percent less than the fourth quarter of 2017, according to the Semiconductor Industry Association (SIA). Sales for the month of March 2018 came in at $37.0 billion, an increase of 20 percent compared to the March 2017 total of $30.8 billion and 0.7 percent more than the February 2018 total of $36.8 billion.

"The global semiconductor market has demonstrated impressive growth through the first quarter of 2018, far exceeding sales through the same point in 2017, which was a record year for semiconductor revenues," said John Neuffer, president and CEO, Semiconductor Industry Association. "Sales in March increased year-to-year for the 20th consecutive month. All regional markets experienced double-digit growth compared to last year, and all major semiconductor product categories experienced year-to-year growth, with memory products continuing to lead the way."

Friday, April 27, 2018

Mediatek halts shipments to ZTE

Following an order from Taiwan's Bureau of Foreign Trade, Mediatek has suspended shipment of its chips and components to ZTE, according to Nikkei Asian Review.

Mediatek is the second largest global supplier of systems-on-chip (SoC) solutions for mobile devices. It also supplies a range of connectivity chips for home networking and broadband CPE, along with a new line of optical components.

https://s.nikkei.com/2HFl4q4

Sunday, April 22, 2018

Alibaba acquires C-SKY Microsystems for embedded IoT CPUs

Alibaba has acquired Hangzhou Zhongtian Microsystem Co. (C-SKY Microsystems, a leading developer of embedded CPUs, for an undisclosed sum.

C-SKY, which was founded in 2001 and is based in Hangzhou's Hi-Tech Zone, offers a series of 32-bit "C-SKY" embedded CPU cores based on independent intellectual property rights. The chips are widely used in Internet of Things intelligent hardware, digital audio and video, information security, networking and communications, industrial control, and automotive electronics.


Friday, April 20, 2018

Intel cuts water usage in semiconductor manufacturing

Intel is investing in a massive water recycling facility at its plant in Hillsboro, Oregon as part of a corporate goal to return 100 percent of water used in its global manufacturing to communities and watersheds for local use by 2025.

When complete, the huge water recycling plant in Hillsboro, will be able to recycle about 1 billion gallons of water every year – the equivalent of 90,000 Olympic-size swimming pools.

See also