Tuesday, July 24, 2018

Google builds its Cloud Services Platform

Google outlined its vision for integrated cloud services running over on-premise equipment as well as its own, massively-scalable cloud infrastructure. At is Google Next conference in San Francisco, Google presented its Cloud Services Platform, which is architecturally aligned with the joint hybrid cloud products being developed with Cisco.

So far, the Cloud Services Platform includes the following:


  • Service mesh: Google is announcing an Istio service for managing services within a Kubernetes Engine cluster. Google is releasing Istio 1.0 in open source, Managed Istio, and Apigee API Management for Istio.
  • Hybrid computing: Google Kubernetes Engine (GKE) On-Prem will provide multi-cluster management
  • Policy enforcement: GKE Policy Management, to take control of Kubernetes workloads
  • Ops tooling: Stackdriver Service Monitoring
  • Serverless computing: GKE Serverless add-on and Knative, an open source serverless framework. Essentially, Google is rolling out serverless containers, which allow customers to run container-based workloads in a fully managed environment and to pay only for actual usage. 
  • Developer tools: Cloud Build, a fully managed CI/CD platform
Google is also rolling out Cloud Functions, an event-driven compute service with an SLA and support for Python 3.7 and Node.js 8, networking and security controls. Cloud Functions can be tied into more than 20 GCP services such as BigQuery, Cloud Pub/Sub, machine learning APIs, G Suite, Google Assistant, etc.

Google is introducing its third-generation cloud tensorflow processing units (TPU)

Google is rapidly expanding its Cloud AutoML (machine learning) capabilities, which is now in public beta testing.


Cisco and Google Partner on New Hybrid Cloud Solution

Cisco and Google Cloud have formed a partnership to deliver a hybrid cloud solutions that enables applications and services to be deployed, managed and secured across on-premises environments and Google Cloud Platform. The pilot implementations are expected to be launched early next year, with commercial rollout later in 2018.

The main idea is to deliver a consistent Kubernetes environment for both on-premises Cisco Private Cloud Infrastructure and Google’s managed Kubernetes service, Google Container Engine.

The companies said their open hybrid cloud offering will provide enterprises with a way to run, secure and monitor workloads, thus enabling them to optimize their existing investments, plan their cloud migration at their own pace and avoid vendor lock in.

Cisco and Google Cloud hybrid solution highlights:


  • Orchestration and Management – Policy-based Kubernetes orchestration and lifecycle management of resources, applications and services across hybrid environments
  • Networking – Extend network policy and configurations to multiple on-premises and cloud environments
  • Security – Extend Security policy and monitor applications behavior
  • Visibility and Control – Real-time network and application performance monitoring and automation
  • Cloud-ready Infrastructure – Hyperconverged platform supporting existing application and cloud-native Kubernetes environments
  • Service Management with Istio – Open-source solution provides a uniform way to connect, secure, manage and monitor microservices
  • API Management – Google's Apigee enterprise-class API management enables legacy workloads running on premises to connect to the cloud through APIs
  • Developer Ready – Cisco's DevNet Developer Center provides tools and resources for cloud and enterprise developers to code in hybrid environments
  • Support – Joint coordinated technical support for the solution

Juniper pushes ahead with 400G based on its own silicon

Juniper Networks is bringing 400GbE capabilities across its wide-area network, data center and enterprise portfolio starting in the second half of 2018, including 400GbE in backbone, peering, data center interconnect, scale-out metro core, telco-cloud services and hyperscale data center IP fabrics.

The announcement includes product refreshes for Juniper’s PTX, QFX and MX series platforms based on the company's recently announced 400GbE-capable Juniper Penta Silicon, and new generations of ExpressPlus and Q5 silicon. Juniper's in-house silicon brings native 400GbE and Flex Ethernet support with tighter MACsec integration. Juniper said its next-generation silicon also brings forward all existing functionality in previous generations while adding deep telemetry, SPRINGv6 and VXLAN EVPN support.

“The impending wave of network traffic is coming from all angles and affecting all industries. Success will be about not only equipping the network with the right technology to handle the traffic in a secure way but also fine-tuning the economics so it makes good business sense. We’re putting a stake in the ground today in leading our customers’ transition to 400GbE network capacity with the industry’s most comprehensive set of use cases, including backbone, peering, data center IP fabric and metro core, to enable our customers to economically usher in the next era of connectivity,” stated Bikash Koley, Chief Technology Officer, Juniper Networks.

Highlights of the announcement:

400GbE for IP Transport

Juniper is unveiling a new 3-RU PTX10003 Packet Transport Router for next-generation backbone, peering and data center interconnect applications ready to deploy high-density 100GbE and 400GbE. The new PTX10003 is the industry’s first packet transport router to accommodate universal multi-rate QSFP-DD interface for seamless 100GbE to 400GbE upgrades. It features native MACsec support for 160x100GbE and FlexE support for 32x400GbE interfaces. The Juniper PTX10003 is powered by the next generation of its ExpressPlus silicon. Commercial availability is expected during the second half of 2018.

400GbE for the Data Center

  • Juniper is introducing a QFX10003 data center switch boasting 32x400GbE in a compact 3-RU form factor. The system can scale up to 160x100GbE giving customers investment protection as they transition to 400GbE. The QFX10003, which will be powered by the next generation Q5 silicon, features a deep buffer enabled by Hybrid Memory Cube memory, allowing it to absorb network traffic spikes and reduce application latency across MACsec encrypted 25GbE, 50GbE, and 100GbE environments. QFX10003 is expected to be available during the second half of 2018.


  • Juniper is also introducing new QFX5220 switches powered by the latest generation of merchant silicon and offering 32x400GbE in a compact 1-RU form factor. These switches also offer flexible 50GbE, 100GbE and 400GbE interfaces for server and inter-fabric connectivity, providing deployment versatility and network investment for the 400GbE cloud transition. The QFX5220 is expected to be available in the first half of 2019.

400GbE for WAN services

  • Juniper's recently announced MX Series 5G Universal Routing Platform, powered by its own new Penta Silicon, offers 400GbE interfaces and industry-first support of native MACsec and IPsec encryption that can originate and terminate thousands of IPsec sessions without sacrificing performance. Juniper Penta Silicon-powered 400GbE MPC10E line cards for the MX960, MX480 and MX240 are expected to be available in the first half of 2019. 


Juniper refreshes MX edge routers powered by its Penta silicon

Juniper Networks introduced its new MX Series 5G Universal Routing Platform powered by a new generation of its custom ASICs promising extensive programmability and performance improvements for supporting resource-intensive applications like 5G and secure SD-WAN-based managed services.

The new "Penta" silicon, which leverages 16nm geometry, functions as the packet forwarding engine for the MX series. It delivers a 50 percent power efficiency gain (0.5 watts per gigabit) over the existing Junos Trio chipset, which leads to a 3x bandwidth increase for the MX960, MX480 and MX240. Juniper designed the Penta silicon with native support of both MACsec and an IPsec crypto engine – an industry-first – that can originate and terminate thousands of IPSec sessions without sacrificing performance. Additionally, Juniper Penta supports flexible native Ethernet support (FlexE).

Juniper Penta silicon features:

  • 16nm with Integrated 3D HBM Memory reducing packaging by 83% over the 4th Generation Trio silicon 
  • 500G Full Duplex / 1T Half Duplex 
  • 50Gbps SERDES 
  • Fully Programmable Packet Pipeline 
  • Fully Fungible Integrated Databases for FIB, ACLs, Tunnels, Telemetry, and more… 
  • Built-in FlexEthernet Support 
  • Built-in MACSec & IPSec Tunnel Encryption Engine (Industry First)
A second major innovation with the platform is open hardware-accelerated 5G Control and User Plane Separation (CUPS), which enables operators to separate the evolved packet core user plane (GTP-U) and control plane (GTP-C) with a standardized Sx interface. This allows service providers to scale the control plane and the user plane independently as needed for added flexibility and investment protection.

Juniper said its MX Series 5G platform is the first networking platform to support a standards-based hardware accelerated 5G user-plane in both existing and future MX routers to enable converged services (wireless and wireline) on the same platform while also allowing integration with third-party 5G control planes. Juniper expects this will lower total cost of ownership by as much as three to four times over software-based user plane implementations for MXs deployed in the field.

In addition, the rollout expands Juniper Universal Chassis system. The previously announced PTX and QFX Universal Chassis gains two new MX variants: MX10008 and MX10016. The 13-RU MX10008 and 21-RU MX10016 Universal Chassis bring industry-leading and space-saving scale for edge routers at 19.2Tbps and 38.4Tbps, respectively.

The MX10008 and MX100016 will be available during the second half of 2018. Juniper Penta Silicon-powered line cards for the MX960, MX480 and MX240 will be available in Q1 2019. New CUPS support will be available in the first half of 2019.

Intel and SiTime collaborate on MEMS timing for 5G

SiTime, which specializes in MEMS timing solutions, and Intel agreed to collaborate on integrating timing solutions for Intel’s 5G multi-mode radio modems, with additional applicability to Intel LTE, millimeter-wave wireless, Wi-Fi, Bluetooth, and GNSS solutions.

SiTime said its MEMS timing technology helps meet the high-performance requirements of emerging 5G radio modem platforms, especially in the presence of stressors such as vibration, high temperature, and rapid thermal transients. Such stressors can disrupt the timing signal and result in network reliability issues, lower data throughput, and even connectivity drops.

“Our collaboration with SiTime on MEMS-based silicon timing solutions will help our customers build leading 5G platforms to best take advantage of the increased performance and capacity that the 5G NR standard brings,” said Dr. Cormac Conroy, corporate vice president and general manager of the Communication and Devices Group at Intel Corporation. “Intel’s modem technology and our collaboration with SiTime is helping to enable new mobile and consumer experiences, and enterprise and industrial use cases.”

“Intel is already building 5G’s future and has the scale to meet 5G’s scope,” said Rajesh Vashist, CEO at SiTime. “Our collaboration enables SiTime to align our MEMS timing solutions roadmap with Intel’s 5G platforms. Intel’s expertise in 5G modems, with SiTime’s game-changing timing technology, is a potent partnership for future growth and one that enables successful deployment of 5G. As SiTime continues to lead the world in MEMS timing, the opportunities between our companies are growing and this agreement sets us both on a path for continued success.”

Verizon to bring 5G fixed wireless access to Houston

In a challenge to Comcast's Xfinity and AT&T's U-verse broadband and video residential services, Verizon will launch a 5G fixed wireless access service in Houston before the end of the year.

Verizon has previously announced Sacramento and Los Angeles as launch cities for its residential 5G broadband service.

Verizon CEO Lowell McAdam, CTO and incoming CEO Hans Vestberg are hosting a news conference with Houston mayor Sylvester Turner.

“We expect 5G will be a game changer helping us usher in a new wave of progress and innovation,” said Mayor Turner. “We’re delighted to be one of the first cities to bring 5G to our local communities and look forward to continuing our long-standing relationship with Verizon.”

“We will be the first to offer commercial 5G service and our work with Houston put us one step closer to delivering that promise,” said Verizon’s Vestberg. “With 5G we are ushering in a fourth industrial revolution that will help reshape cities and lead to unprecedented innovation, and Houston will be at the forefront of that innovation.”

Verizon readies 5G fixed residential service launch in 2018

Verizon's first 5G application will be a fixed residential broadband service.

The first commercial launch will occur in Sacramento, California during the second half of 2018. Additional markets are also expected to launch in 2018.

Verizon did not disclose the performance characteristics of its 5G residential service but said it will provide unprecedented wireless speeds for Internet access.

Verizon estimates the market opportunity for initial 5G residential broadband services to be approximately 30 million households nationwide. The company also noted that the 5G residential rollout will not have a material impact on Verizon’s consolidated capital expenditures in 2018. CAPEX is expected to be consistent with the past several years.

“This is a landmark announcement for customers and investors who have been waiting for the 5G future to become a reality,” said Hans Vestberg, Verizon president of Global Networks and Chief Technology Officer. “We appreciate our strong ecosystem partners for their passion and technological support in helping us drive forward with 5G industry standards, for both fixed and mobile applications. The targeted initial launches we are announcing today will provide a strong framework for accelerating 5G’s future deployment on the global standards.”

Verizon picks Samsung as a supplier for Fixed 5G

Verizon has selected Samsung Electronics America as a supplier of commercial 5G Fixed Wireless Access (FWA) network solutions. Verizon will begin by launching commercial 5G services in Sacramento, California in the second half of 2018. Financial terms were not disclosed.

Specifically, Samsung will provide Verizon with commercial 5G home routers (CPEs), 5G Radio Access Units (RAN) comprised of a compact radio base station and virtualized RAN elements, as well as 5G radio frequency planning services. Samsung said it has leveraged in-house technology and assets to develop the first commercial ASIC-based 5G modems and mmWave RFICs.

Samsung and Verizon began 5G customer trials across seven U.S. cities in early 2017, and have successfully tested and verified 5G performance using millimeter-wave (mmWave) frequency to provide FWA pre-commercial service. These trials were conducted in California, Georgia, New Jersey, Massachusetts, Michigan, Texas, and Washington D.C.

The companies cited the following findings from their 5G testing:

  • A single 5G radio has been able to reach the 19th floor of a multi-dwelling unit (MDU).
  • Broadband service has been achieved in line of sight (LOS), partial LOS and Non-LOS connections.
  • Environmental factors, including rain and snow, have not interrupted services.

“The industry has been discussing 5G connectivity for years, and through our joint collaboration with partners like Samsung, we are beginning to make it a reality for our customers," said Ed Chan, Chief Technology Architect and Network Planning, Verizon. 

Verizon has 116.5 million retail wireless connections

Verizon reported consolidated operating revenues of $32.2 billion for Q2 2018, up 5.4 percent from second-quarter 2017. On a comparable basis excluding the impacts of Oath, divested businesses and the revenue recognition standard (non-GAAP), consolidated revenues were $30.2 billion, up approximately 2.6 percent.  Net income was $4.2 billion and EBITDA (non-GAAP) totaled approximately $11 billion. EPS was $1.00, compared with $1.07 in second-quarter 2017.

“Verizon is extremely well-positioned for the future,” said Chairman and CEO Lowell McAdam. “Our financial and operating results for the first half of 2018 were strong, as evidenced by service revenue, earnings and operating cash flow growth delivered in a highly competitive marketplace.”

Highlights

Wireless

  • Total revenues were $22.4 billion, an increase of 5.5 percent year over year. 
  • Service revenues for the quarter on a reported basis grew 0.8 percent year over year. 
  • Approximately 82 percent of Verizon’s postpaid phone base were on unsubsidized plans, compared with 81 percent in first-quarter 2018 and 75 percent in the same period last year.
  • Net increase of 531,000 retail postpaid additions, consisting of net phone additions of 199,000, tablet losses of 37,000 and 369,000 other connected devices additions, primarily wearables. Postpaid smartphone net additions for the quarter were 398,000.
  • Total retail postpaid churn was 0.97 percent, slightly up year-over-year. Retail postpaid phone churn of 0.75 percent in second-quarter 2018 was the fifth consecutive quarter of retail postpaid phone churn of 0.80 percent or better.

Wireline

  • Total wireline revenues were $7.5 billion. Excluding the impact of the revenue recognition standard, total wirelines revenues decreased 3.4 percent year over year in second-quarter 2018.
  • Total Fios revenues were $3.0 billion, an increase of 2.0 percent year over year. 
  • Added a net of 43,000 Fios Internet connections, indicative of strong demand as customers value broadband connections more than ever. Verizon lost 37,000 Fios Video connections in second-quarter 2018 amid pressures from cord-cutting of video bundles.
  • Wireline operating loss was $19 million in second-quarter 2018, and segment operating loss margin was 0.3 percent.
  • CAPEX for 1H 2018 was $7.8 billion, compared to $7.0 billion for 1H 2017


https://www.verizon.com/about/investors/quarterly-reports/2q-2018-quarter-earnings-conference-call-webcast

AT&T posts declines in domestic video and legacy wireline

AT&T reported solid wireless results in the second quarter, including postpaid phone gains, continued strong prepaid phone growth and stable postpaid churn. On a GAAP basis, service revenue declined; however, on a comparable basis service revenue grew.  Consolidated revenue totaled $39.0 billion versus $39.8 billion in the year-ago.

Declines in domestic video and legacy wireline services were offset by adding approximately $1.1 billion from Time Warner net of eliminations and growth in wireless, strategic business services and advertising. On a comparative basis, revenues were $39.9 billion, an increase of 0.2% primarily due to the second-quarter close of the Time Warner acquisition.

U.S. wireless results:

  • Service revenue growth on a comparable basis
  • 46,000 postpaid phone net adds with continued strong year-over-year improvement
  • Continued prepaid growth with 356,000 phone net adds
  • Nearly 400,000 branded smartphones added to base
  • Second-quarter postpaid phone churn of 0.82%

Entertainment Group results:

  • 342,000 DIRECTV NOW net adds to reach more than 1.8 million subscribers
  • 80,000 total video net adds; total video customer base stable with DIRECTV NOW; AT&T WatchTV launched
  • 76,000 IP broadband net adds; 23,000 total broadband net adds; more than 9 million customer locations passed with fiber
  • AdWorks continues double-digit revenue growth
  • Time Warner acquisition closed on June 14; full second-quarter results include:
  • HBO and Turner year-over-year subscription revenue growth
  • Turner ad revenues up 3%
  • Record number of series in production at Warner Bros.
  • 166 Primetime Emmy Awards nominations

AT&T also issued the following updated 2018 guidance:

  • Raising adjusted EPS to high end of the $3.50 range
  • Raising free cash flow to high end of the $21 billion range; inclusive of all deal and integration costs
  • Capital Investment of approximately $25 billion; $22 billion net of expected FirstNet reimbursements and vendor financing

Openreach sets wholesale pricing for UK fibre broadband

Openreach announced a new, supplementary wholesale discount structure for fibre broadband in the UK in return for volume commitments.

Openreach said it would offer long-term discounts to all of its Communication Provider (CP) customers from 21 August. The discounts go beyond Ofcom’s pricing controls on its superfast broadband products.

BT believes that the new pricing structure will allow CPs to encourage more of their customers onto better services and ultimately to move the vast majority of Britain’s homes and businesses onto superfast and ultrafast platforms.

The Openreach pricing tiers are posted here:

https://www.openreach.co.uk/orpg/home/products/super-fastfibreaccess/downloads/Openreach_Special_Offer_GEA_Volume_Agreement.pdf

UNH-IOL appoints Jeff Lapak as executive director

The University of New Hampshire InterOperability Laboratory (UNH-IOL) named Jeff Lapak as executive director of the UNH-IOL.

Michayla Newcombe has been promoted to associate director of UNH-IOL.

Before this appointment, Jeff has held the positions of associate director, industry lead, and senior manager at the UNH-IOL. As a leader, he’s helped oversee development for many physical layer testing technologies such as 10, 25, 40, and 100 Gigabit Ethernet. Jeff began working at the UNH-IOL as a UNH electrical engineering undergraduate student in 1998 and continued on to earn his MBA from UNH in 2013.

Michayla started her career at the UNH-IOL in 2012 as a DHCPv6 test technician while she was an undergraduate at UNH. She graduated from UNH in 2014 with a bachelor's degree in Business Administration and continued to work full-time at the UNH-IOL as quality manager for ISO-17025 accreditation. As the manager for the IP industry, Michayla has recently helped drive the USGv6 and IPv6 Ready Logo Test Programs’ newest programs. With the lab’s recent switch to Agile Methodologies, Michayla leads several agile teams within the UNH-IOL including SDN, IOT and INTACT tool development.

https://www.iol.unh.edu/

Monday, July 23, 2018

Infinera to acquire Coriant for $430 million

Infinera agreed to acquire privately-held Coriant, creating one of the world’s largest optical network equipment providers. Infinera will pay approximately $150 million in cash at closing, and estimated additional amounts of $25 million in the two quarters post-closing and $55 million over a period of years. Infinera will issue approximately 21 million shares, which when combined with the cash consideration, results in total transaction consideration of approximately $430 million. Coriant’s majority owner, Oaktree Capital Management, will be compensated largely in stock. Post-transaction, Oaktree will own approximately 12 percent of the combined company on a fully diluted basis.

Infinera said the acquisition significantly scales the company as the next wave of global network spending begins as operators transform their networks from 4G to 5G, from Optical Transport Network (OTN) to packet and from closed to open network architectures. The company also says its vertically integrated business model delivers higher performance at lower cost for customers and higher profit margins for stockholders. The acquisition approximately doubles Infinera’s revenue and expands the company’s customer base to serve nine of the top 10 global network operators (five new to Infinera) and the top six global internet content providers (three new to Infinera).

“Acquiring Coriant is a fantastic opportunity, strengthening our ability to serve the world’s largest network operators, accelerating our ability to leverage vertical integration and reinforcing our commitment to our long-term business model,” said Tom Fallon, Infinera CEO. “This powerful combination immediately benefits our combined customers by delivering the innovative technology required for the next wave of network spending.”

“Joining forces with Infinera accelerates our vision of transforming next-generation networks with best-in-class innovation,” said Pat DiPietro, CEO and Vice Chairman of Coriant. “The combined company will benefit from Coriant’s significant investments in software automation and open, disaggregated solutions, which have gained significant traction and wide-scale adoption by customers around the world.”

The companies also noted:

  • Coriant invested nearly $1 billion in research and development over the last five years to significantly advance its portfolio to address the growing demand for software automation and open, disaggregated platforms. 
  • Approximately 70% of Coriant's product revenue in 2017 (~80% projected in 2018) came from solutions that are new or have been refreshed in the last three years. 
  • Coriant has over 80 trials in progress or have been completed to date in 2018 for Coriant’s suite of new products and automation tools. 
  • Coriant’s SDN orchestrator Maestro has been deployed operationally at one of the two largest Tier 1 service providers in North America. 
  • Coriant has added more than 20 new customers to date in 2018, seven of which are driving more than $1 million in annual revenue.
  • Infinera recently upgraded its portfolio based on its Infinite Capacity Engine (ICE4). 
  • The acquisition is expected to close in the third quarter of 2018, subject to customary closing conditions. 

Coriant was formed in 2013 by private equity firm Marlin Equity Partners, combining Nokia Siemens Networks, Tellabs and Sycamore Networks.

http://investors.infinera.com/home/default.aspx



Coriant announces CEO change

Coriant named Pat DiPietro as its new Chief Executive Officer, effective immediately, replacing CEO and Chairman Shaygan Kheradpir, who has stepped down from his role to pursue other opportunities.

DiPietro will continue to serve as Vice Chairman of the Board, a role he has held since the founding of Coriant in 2013. DiPietro previously held senior leadership roles at Nortel and Bell Northern Research. He also previously served as Managing Partner at Canada’s VG Partners, overseeing the company’s Technology Fund. As a venture capitalist, he managed large portfolios and teams and sat on numerous Boards, including Sandvine, SiGe, Continuous Computing, BTI Systems and BelAir Networks.

Coriant also announced that Reza Ghaffari has been promoted to the role of Chief Operating Officer (COO), a new position within Coriant. Ghaffari will continue to lead Coriant’s global service and support organization, while assuming responsibility for the company’s global IT, human resources, and facilities functions. Between 2000 and 2005, Ghaffari also worked at Verizon where he was responsible for innovation, product development, and strategic partnership programs.

Singtel plans 5G pilot for Q4

Singtel expects to kickoff its 5G trial by the fourth quarter of this year. The 5G pilot network, which will be conducted in partnership with Ericsson, will occr a0t one-north, the country’s science, business and IT hub. Singapore's Info-Communications Media Development Authority (IMDA) has allocated spectrum for the trial.

“5G has the potential to accelerate the digital transformation of industries, as well as empower consumers with innovative applications,” said Mr Mark Chong, Group Chief Technology Officer at Singtel. “We are pleased to take another bold step in our journey to 5G with our 5G pilot network at one-north and invite enterprises to start shaping their digital future with us.”

“This is an encouraging step towards commercialisation with live 5G trial networks made possible with the regulatory sandbox IMDA has in place,” said Ms Aileen Chia, Deputy Chief Executive and Director-General (Telecoms & Post) at IMDA.  “IMDA will continue to work closely with mobile service providers such as Singtel in their journey to build communication capabilities of the future and complement Singapore’s efforts towards a vibrant digital economy.”

At a “Bringing 5G to Life” event held at Singtel Comcentre this week, Singtel and Ericsson demonstrated c3D augmented reality (AR) streaming over a 5G network operating in the 28GHz millimetre wave spectrum. The event also showcased emerging technologies from partners – Garuda Robotics, Intel, Meta, Rohde & Schwarz and Sony.

Ericsson wins more Verizon LTE markets

Verizon has selected Ericsson to supply 4G equipment in additional markets across the U.S. Financial terms were not disclosed.

The installations include Ericsson Radio System 4G LTE equipment, which is a 5G-ready platform that will allow Verizon to rapidly transition sites to 5G when they deploy the service in those markets. Specifically, Ericsson said the new markets will be deployed using its latest baseband and dual band radios, an optimized solution for widescale high-capacity deployment. The solution will be software upgradeable to 5G and will deliver CAT-M and NB-IoT for Massive Internet of Things use cases, as well as industry-leading network capacity.

Verizon and Ericsson cited a number of recent joint milestones:

  • Achieved 953 Mbps by combining licensed and unlicensed bands with four-carrier aggregation, 4x4 MIMO and 256 QAM in a real-world network environment in Florida
  • Exceeded 1 Gbps speeds on the Verizon 4G LTE network using three-carrier aggregation
  • Verizon and Ericsson reached another milestone in 4G LTE-Advanced technologies that will serve as a stepping stone to 5G by completing their first deployment of FDD (Frequency Division Duplexing) Massive MIMO (Multiple Input – Multiple Output) on Verizon’s wireless network in Irvine, CA.

Yossi Cohen, Head of Customer Unit Verizon, Ericsson North America, says: “We continue to work closely with Verizon to expand the strategic partnership between our companies, enabling Verizon to unleash the latest in 4G technology and simultaneously prepare to launch 5G in these new markets.”

Qualcomm delivers 5G NR mmWave antenna modules

Qualcomm Technologies has begun sampling fully-integrated 5G NR millimeter wave (mmWave) and sub-6 GHz RF modules for smartphones and other mobile devices.

The Qualcomm QTM052 mmWave antenna module family and the Qualcomm QPM56xx sub-6 GHz RF module family pair with the Qualcomm Snapdragon X50 5G modem to deliver modem-to-antenna capabilities across several spectrum bands, in a compact footprint that is suited for integration in mobile devices.

“Today’s announcement of the first commercial 5G NR mmWave antenna modules and sub-6 GHz RF modules for smartphones and other mobile devices represents a major milestone for the mobile industry. Qualcomm Technologies’ early investment in 5G has allowed us to deliver to the industry a working mobile mmWave solution that was previously thought unattainable, as well as a fully-integrated sub-6 GHz RF solution. Now, these type of modem-to-antenna solutions, spanning both mmWave and sub-6 spectrum bands, make mobile 5G networks and devices, especially smartphones, ready for large scale commercialization,” said Cristiano Amon, President, Qualcomm Incorporated. “With 5G, consumers can expect gigabit-class Internet speeds with unprecedented responsiveness in the palm of their hands, which stand to revolutionize the mobile experience.”

About the modules

  • The QTM052 mmWave antenna modules work with the Snapdragon X50 5G modem. They support advanced beam forming, beam steering, and beam tracking technologies to improve the range and reliability of mmWave signals. They feature an integrated 5G NR radio transceiver, power management IC, RF front-end components and phased antenna array. They support up to 800 MHz of bandwidth in the 26.5-29.5 GHz (n257), as well as the entire 27.5-28.35 GHz (n261) and 37-40 GHz (n260) mmWave bands.  Qualcomm says that up to QTM052 modules can be integrated in a smartphone, enabling OEMs to continue evolving the industrial design of their mobile devices, offering attractive form factors combined with the benefits of extremely high speeds from mmWave 5G NR, and making such devices available for launch as early as the first half of 2019. 
  • The QPM56xx RF module family (including the QPM5650, QPM5651, QDM5650, and QDM5652) is designed to allow smartphones based on the Snapdragon X50 5G modem to support 5G NR in sub-6 GHz RF bands. The QPM5650 and QPM5651 feature integrated 5G NR PA/LNA/Switch and filtering subsystem. The QDM5650 and QDM5652 feature integrated 5G NR LNA/switch and filtering subsystem for diversity and MIMO support. All four modules offer integrated SRS switching required for optimum massive MIMO applications and support for 3.3-4.2 GHz (n77), 3.3-3.8 GHz (n78) and 4.4-5.0 GHz (n79) sub-6 bands. These sub-6 GHz RF modules provide mobile device makers with a viable path to delivering on the promise of 5G NR massive MIMO technology in mobile devices.


Iron Mountain ties into Google Cloud as preferred provider

Iron Mountain, which specializes in secure corporate storage solutions, is working with Google to develop AI-powered SaaS solutions on Google Cloud Platform that help organizations analyze their vast physical and digital information and data repositories, to unlock insights, improve decision making and create new revenue streams.

Iron Mountain expects to deliver these new subscription services built on GCP this September.

In addition, Iron Mountain has selected Google Cloud as its preferred cloud provider for artificial intelligence and machine learning.

"As the trusted guardian of information for 95 percent of the FORTUNE 1000, we know firsthand the challenges of understanding and acting on data insights," said William Meaney, president & CEO, Iron Mountain. "Whether it's a global insurance company or an oil and gas conglomerate, our customers depend on us to help make their physical and digital information useful and accessible while keeping it safe. This partnership with Google Cloud brings two industry-leading players together to deliver a new category of solutions for organizations looking to utilize their data's value, while closely managing their security and privacy."

"We're delighted to partner with Iron Mountain to develop and bring new insight capabilities to market - particularly given their deep domain expertise and customer portfolio," said Tariq Shaukat, President, Partner and Industry Platforms at Google Cloud. "Customers in data-intensive industries like energy, financial services, healthcare and many others can create significant business value from the union of data with machine learning and content analytics."


Arista'a Any Cloud hybrid platform reaches Google Cloud Platform

Arista's Any Cloud hybrid cloud networking platform is now available on Google Cloud Platform. This includes the virtualized Arista vEOS Router combined with CloudVision with Cloud Tracer functionality.

Arista's vEOS Router leverages a cloud-grade routing stack and IPsec VPN tunnels to reliably and securely interconnect workloads across multi-cloud deployments. In addition, vEOS Router is integrated with Google Cloud provider APIs to enable easy deployment, automation, analytics, and end-to-end visibility. Arista CloudVision provides advanced visibility and automation, extending the cloud networking model of Arista on-premises cloud network platforms to Google Cloud Platform.

Arista said its hybrid cloud platform now offers cloud platform-specific images for Amazon Web Services, Google Cloud Platform, Microsoft Azure Cloud, Microsoft Azure Stack on-premise Cloud, and common on-premises hypervisor platforms. Support in each environment is coupled with validation and registration of these solutions in the cloud marketplace infrastructure provided by each cloud provider, thus making deployment simple for enterprise customers.

Arista's Any Cloud aims for universal cloud consistency

Arista Networks unveiled its strategy for extending its Extensible Operating System (EOS) across private cloud datacenters and public cloud providers.

The Arista Any Cloud software platform will support Amazon Web Services (AWS), the Microsoft Azure cloud platform, Microsoft Azure Stack, Google Cloud Platform and Oracle Cloud Infrastructure.  Arista will also be providing integration with the Equinix Cloud Exchange, which provides direct high-performance connections to 70+ cloud providers.

Key elements of the Arista Any Cloud solution include the Arista vEOS Router for hypervisors (KVM and VMware ESX) and for cloud platforms (AWS and Azure), and the Arista Cloud Tracer as a CloudVision-based application. Arista’s new Cloud Tracer assures that visibility and availability metrics are tracked consistently across the entire hybrid cloud environment, including public cloud direct connections, remote datacenter connections and cloud exchange points.

Arista said its Any Cloud solution will provide its enterprise customers with a common Universal Cloud Network experience.

“Arista’s Any Cloud platform transcends public and private clouds, radically changing the on-premises enterprise datacenter. We are helping customers realize their hybrid-cloud transformation by extending Arista EOS and CloudVision across network boundaries,” said Jayshree Ullal, President and CEO for Arista Networks.

Google posts another strong quarter with sales up 26%

Alphabet reported revenue of $32.657 billion for its Q2 ended 30-June-2018 up 26% versus the second quarter of 2017 and 23% on a constant currency basis.

"We delivered another quarter of very strong performance, with revenues of $32.7 billion, up 26% versus the second quarter of 2017 and 23% on a constant currency basis. Our investments are driving great experiences for users, strong results for advertisers, and new business opportunities for Google and Alphabet," said Ruth Porat, CFO of Alphabet and Google

Digital Realty establishes direct connect to IBM Cloud in 15 cities

Digital Realty announced dedicated, private access to the IBM Cloud in 15 major metropolitan areas around the world.  These

Direct access to IBM Cloud is now available through the Digital Realty Service Exchange and IBM Cloud Direct Link via more than 70 Digital Realty data centers in Amsterdam, Ashburn, Atlanta, Boston, Chicago, Dallas, Frankfurt, Hong Kong, London, Los Angeles, Melbourne, Miami, Phoenix, Portland, New York, Santa Clara, Seattle, Singapore, Sydney and Toronto.

"Businesses are turning to the cloud to extract new value from their data and deliver better customer experiences faster than ever before," said Kit Linton, Vice President of Network, IBM Cloud.  "The combination of Digital Realty's extensive global reach and highly resilient connectivity from IBM Cloud Direct Link can help businesses build the protected, hybrid cloud environment they need to innovate at scale."

"Providing enterprise connectivity to the extensive range of services available through IBM Cloud via Direct Link ensures customers not only have access to the compute and storage resources they need, but they are also able to leverage more advanced cloud applications like analytics, blockchain technology and the Internet of Things," said Digital Realty Chief Technology Officer Chris Sharp.  "We look forward to continuing to work closely with IBM to enable our mutual customers to access the cloud services they need to accelerate their own business growth, no matter where they may be located."

Toshiba ships NVMe SSDs based on 3D Flash

Toshiba Memory Corporation launched its first NVM Express SSDs utilizing 96-layer, fourth generation, 3D BiCS FLASH memory technology.

The new XG6 series client SSDs features a PCI Express (PCIe) Gen3 x 4Lane and NVM Express (NVMeTM) 1.3a interface. The XG6 series delivers sequential write performance of up to 2960 MB/s in client SSDs. Sequential read performance is up to 3180 MB/s, up to 355,000 random read and 365,000 random write IOPS. The power consumption is maximum 4.7W in active mode and 3mW during stand-by in the lowest power mode, which makes it suitable for power-sensitive mobile PCs.

The new SSDs will be available in three capacities, 256GB, 512GB and 1024GB, all on a single-sided M.2 2280-S2 form factor.

Sunday, July 22, 2018

Stanford researchers train neural network using optical chip

Researchers at Stanford University have demonstrated the ability to train artificial neural networks directly on an optical chip.

Neural network processing currently relies on conventional computers.  The Optical Society's journal for high impact research, Stanford University researchers report a method for training these networks directly in the device by implementing an optical analogue of the ‘backpropagation’ algorithm, which is the standard way to train conventional neural networks.

“Using an optical chip to perform neural network computations more efficiently than is possible with digital computers could allow more complex problems to be solved,” said research team leader Shanhui Fan of Stanford University. “This would enhance the capability of artificial neural networks to perform tasks required for self-driving cars or to formulate an appropriate response to a spoken question, for example. It could also improve our lives in ways we can’t imagine now.”

https://www.osapublishing.org/optica/abstract.cfm?uri=optica-5-7-864