Networking trends for 2018

Here are some of my collection of newest trends and predictions for year 2018. I have not invented those ideas what will happen next year completely myself. I have gone through many articles that have given predictions for year 2018. Then I have picked and mixed here the best part from those articles (sources listed on the end of posting) with some of my own additions to make this posting.This article contains very many quotations from those source articles.

Big Data, mobility and the Internet of Things (IoT) are generating an enormous amount of data, and data center operators must find ways to support higher and higher speeds. Recent data center trends predict bandwidth requirements will continue growing 25 percent to 35 percent per year. Many older data centers were designed to support 1-gigabit or 10-gigabit pathways between servers, routers and switches. Today’s Ethernet roadmap extends from 25- and 40-gigabit up through 100-gigabit, and 400-gigabit and even 1-terabit Ethernet loom within a few years. The biggest sales are forecasted for 25G and 100G ports for next few years. Ethernet switch market has now 100 Gbit/s products in the volumes at the moment, and both 200G and 400G Ethernet versions will be taken to use in 2018.

Broadcom dominates the Ethernet switch chip market today with a 73 to 94 percent share, depending on how market watchers slice the sector valued at nearly a billion dollars. Its closest rival, Cisco Systems, takes most of the rest with systems using its own ASICs. Juniper, Hewlett Packard Enterprise and Huawei also make Ethernet switch ASICs for their systems. With seven merchant chips in the pipeline and four in-house ASICs in the works, we will see a record number of unique platforms ship in 2018. Wheeler of the Linley Group expects the competition will drive Ethernet switch costs from about $60/port today to about $36/port by 2020.

Data center giants “are driving their own code and programmable capabilities as close to the server as possible.The dozen largest data center operators — including the likes of Facebook and Google — build their own switch systems or specify systems built by ODMs. They can drive sales of millions of chips a year but demand maximum bandwidth at minimum cost and power consumption. The Tomahawk-3 is geared for the next-generation of their top-of-rack and aggregation switches, delivering up to 128 100GE or 32 400GE ports, the first merchant chip to support 400GE rates.

China will start making more optical components: Several Western component and subsystems vendors have cited reduced demand from Chinese systems houses such as Huawei and ZTE for revenue declines in 2017. One reason for the slowdown is the fact that these systems houses have begun looking for more local optical technology sources. In addition to doing more development work in-house (particularly in the case of Huawei), the two Chinese systems vendors have begun to work more closely with Chinese companies such as Accelink, Hisense, and HiSilicon as well as Japanese vendors. This can mean that Western firms (particularly in the U.S.) may not see their Chinese orders return to previous levels.

Higher power power over Ethernet: 802.3bt – IEEE Draft Standard for Ethernet Amendment: Physical Layer and Management Parameters for DTE Power via MDI over 4-Pair amendment to IEEE Std 802.3-2015 increases the maximum PD power available by utilizing all four pairs in the specified structured wiring plant. This represents a substantial increase to the capabilities of Ethernet with standardized power – allow delivery of up to 90 watts of power via existing Ethernet cabling. The Ethernet Alliance has announced details of its next plugfest: Dedicated to pre-standard testing of Power over Ethernet (PoE) technologies against Draft 3.2 of the IEEE P802.3bt standard, the event will be held in February 2018. The specification’s ratification is expected in September of 2018.

802.11ax hasn’t been signed off yet, but promises to send WiFi towards 10 Gb/s thanks to its use of both multi-user multiple-input and multiple-output (MU-MIMO) and the new Orthogonal frequency-division multiple access (OFDMA). 802.11ax is  good at combining lots of different links so that users get more connections, more often, and end up with more bandwidth.  Marvell claims it’ll have the first chipsets for new 10G WiFi ready for products in H2 2018. Marvell said the chipsets will ship some time in early 2018 and will appear in products in the second half of the year. Widespread 802.11ax adoption in devices probably won’t happen until 2019.

5G something in it for everyone. 5G is big.  5G New Radio (NR) wireless technology will ultimately impact everyone in the electronics and telecommunications industries. Most estimates say 2020 is when we will ultimately see some real 5G deployments on a scale. In the meantime, companies are firming up their plans for whatever 5G products and services they will offer. Though test and measurement solutions will be key in the commercialization cycle. 5G is set to disrupt test processes. If 5G takes off, the technology will propel the development of new chips in both the infrastructure and the handset. Data centers require specialty semiconductors from power management to high-speed optical fiber front-ends. 5G systems will drive more complexity in RF front-ends .

Networks will become more and more virtual, especially on 5G. 5G networks will build on LTE network architecture with the introduction of cloud RANs (C-RANs) and virtualized RANs. Network function virtualization (NFV) and software-defined networking (SDN) tools and architectures could enable operators to reduce network costs and simplify deployment. For more details read System architecture milestone of 5G Phase 1 is achieved article.

Automotive Ethernet: Automotive Ethernet will replace the Media Oriented Systems Transport (MOST) bus found on many vehicles today and also compete with systems like Maxim’s Gigabit Multimedia Serial Link (GMSL). The standards include IEEE 802.3bw 100BASE-T1 and IEEE P802.3bp 1000BASE-T1 that adopt the 100-Mb/s and 1-Gb/s Ethernet protocols to run over over a single twisted pair up to 15 meters. Chips, PHYs and switches are now readily available for automotive Ethernet. Automotive switches will support time-sensitive networking (TSN) features like audio video bridging (AVB) ingress policy, rate limiting andalso features 802.1Qav/Qbv queue-shaping support. We will need a Security Blanket for Automotive Ethernet.

Car-to-car communication: The industry and government have defined several versions of vehicular communications. These are vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), and vehicle-to-everything (V2X). V2V is the direct communications between vehicles within range of one another. V2V link-up is automatic and cars exchange status information messages Basic Safety Message (BSM) 10 times per second (contains data such as GPS location, speed, direction, brake status, and related conditions). There are competing technologies to make this all work. IEEE802.11p ahead of LTE-V2V for safety critical applications. 5G will offer increased capacity and decreased latency for some critical applications such as vehicle-to-vehicle (V2V) or vehicle-to-infrastructure (V2I) communications for advanced driver assistance systems (ADAS) and self-driving vehicles. The big question is whether 5G will disrupt the landscape or fall short of its promises.

TSN (time sensitive networking) will be talked on many application. Success in the IIoT requires that information- and operational-technology networks work in tandem—time-sensitive networking can make it happen. A converged industrial network will address several challenges that currently exist in today’s disparate network architecture, and it is believed that needed convergence is provided by time-sensitive networking (TSN). TSN is a set of IEEE 802 sub-standards that, when implemented, enable deterministic communication over Ethernet networks while keeping the benefits of communication in best effort manner, on that same network. TSN introduces different traffic classes that share the same link. Proper implementation of TSN requires a solution that can provide a low latency and deterministic response at TSN end points and TSN bridges – it is usually implemented with combining a processor and a FPGA or using advanced Ethernet switch chip.

Gartner analyst says on-premises data centers will soon be ‘useless’: Govekar said that as soon as 2019, at least a third of the largest software vendors will have transitioned their products from cloud-first to cloud-only. For this reason Gartner analyst predicts doom for on-premises data centers. If this happens , there will be a lot of work for network operators and cloud service companies to be able to build the infrastructure that can handle all this without problems. I expect that not all on-premises data centers are doomed. There are applications where moving everything to central could does not make sense – for example many IoT applications are moving some of the processing from cloud to edge devices for various reasons (lower latency, reduce needed bandwidth, work also when communications does not work etc..). There is also place for some on-premises data centers on some application (may them be depending on need anything from embedded server to room full of equipment racks). When designing for the IoT, security needs to be addressed from the Cloud down to each and every edge device. Protecting data is both a hardware and a software requirement, as more data is being stored and analyzed in edge devices and gateways.

Network neutrality talks will continue in 2018. Federal Communications Commission (FCC) on December 14, 2017 overturned the Title II-based Open Internet Order the previous Democratic-controlled FCC regime put in place in 2015. So what happens next? In the immediate term, nothing. The Restoring Internet Freedom Order won’t go into effect until sometime in the first half of 2018.

IPv6 usage seems to be finally accelerating in 2018. IPv6 has been a “future” since 1998, and an important future since 2007. IPv6 deployments have been increasing and chances are you have already used IPv6 – but haven’t realized it yet. IPv6 deployment is increasing around the world, with over 9 million domain names and 23% of all networks advertising IPv6 connectivity. Network admins will have many concerns about migrating to IPv6 in 2018. China plans to put the Internet Protocol version 6 (IPv6)-based network into large-scale use, to boost the development of the Internet industry. China aims to have 200 million active users of IPv6 by the end of 2018. IPv6 use is increasing, but that does not mean that IPv4 is no way dying. It seems that both of those technologies will co-exist in Internet for a long time.

 

Sources:

Chinese systems houses engaging with alternative component, subsystem sources: LightCounting

802.3bt – IEEE Draft Standard for Ethernet Amendment: Physical Layer and Management Parameters for DTE Power via MDI over 4-Pair

Next Ethernet Alliance PoE Plugfest to provide critical support for Power over Ethernet certification program

Making 5G Happen

UK Updates 5G Strategy

System architecture milestone of 5G Phase 1 is achieved

Five technology trends for 2018

NI Trend Watch 2018 explores trends driving the future faster

ADAS Needs V2X to Meet ITS Goals

Taking Automotive Ethernet for a Test Drive

A Security Blanket for Automotive Ethernet

TSN: Converging Networks for a Better Industrial IoT

Gartner analyst says on-premises data centers will soon be ‘useless’

Gartner analyst predicts doom for on-premises data centers

M2M within the IoT – Pushing Security from the Cloud Down to Every Last Endpoint

Net Neutrality Overturned: Now What?

B’com Shifts Switch to 12.8 Tbits/s

Planning starts now for high-speed data center migration

Lights, camera, 802.11ax-ion!

China to speed up IPv6-based Internet development

State of IPv6 Deployment 2017

Top 5 Concerns of Network Admins About Migrating to IPv6 in 2018

 

1,081 Comments

  1. Tomi Engdahl says:

    https://selinc.com/mktg/122222/?utm_source=facebook&utm_medium=social_sdn&utm_campaign=mpn1603&utm_term=sdn&utm_content=sdn_cyber

    Traditional networks use features such as MAC tables, the Rapid Spanning Tree Protocol (RSTP), and cast types for conveniences like “plug and play” functionality. However, these features also make traditional networking vulnerable to cybersecurity threats.

    With software-defined networking (SDN) from SEL, all network flows and backup paths are specifically defined in the controller, so there is no need for MAC tables or RSTP. In addition, SDN uses traffic engineering to process forwarding behavior rather than relying on “cast types,” which pose security risks. This processing eliminates common LAN security threats, including:

    MAC flooding, in which attackers overwhelm the switch with MAC addresses.
    MAC table poisoning, in which attackers convince the switch that they are a false MAC address.
    Address Resolution Protocol (ARP) spoofing, in which an attacker sends false ARP messages, causing traffic to be misdirected.
    Bridge Protocol Data Unit (BPDU) attacks, which disrupt the network’s spanning-tree protocol.
    Flooding using multicast and broadcast Ethernet destinations.

    Reply
  2. Tomi Engdahl says:

    ENGINEER A BETTER NETWORK
    Introducing the industry’s first field-hardened SDN Ethernet switch.
    https://selinc.com/solutions/p/software-defined-network/

    The breakthrough software-defined networking (SDN) technology in the SEL-2740S Software-Defined Network Switch solves the inherent limitations of Ethernet networks. Now, every network path is defined by you; you control how your system responds to network failures. You decide who and what applications are and are not allowed. Instead of programming every device, program one. Instead of waiting on discovery or convergence times, predefine every primary and failover path for the most recoverable Ethernet network possible.

    Reply
  3. Tomi Engdahl says:

    How to relate eye-mask tests to BER
    https://www.edn.com/electronics-blogs/eye-on-standards/4460403/How-to-relate-eye-mask-tests-to-BER?utm_source=Aspencore&utm_medium=EDN&utm_campaign=social

    To perform an eye mask test, engineers used to draw a mask on the display of an analog oscilloscope with a grease pencil. Next, they had to trigger the oscilloscope with clock signal, turn up the persistence, and if the interior of the sketch remained dark, the signal passed. From there, we went to well-defined masks on both real-time and equivalent-time digital oscilloscopes; pixels that lit up within the mask were bad news. Modern mask tests specify a maximum allowed number of these “violations” over a large number of waveforms.

    While eye-mask measurements give you a quick way to gauge if a transmitter is working, they don’t tell you much about the only thing that matters: the bit-error ratio (BER, the number of bit errors to the total number of transmitted bits).

    Reply
  4. Tomi Engdahl says:

    How the Internet of Things Could Fracture Wi-Fi
    New mesh Wi-Fi networks improve coverage, but at the cost of interoperability
    https://spectrum.ieee.org/telecom/wireless/how-the-internet-of-things-could-fracture-wifi

    Wi-Fi is the invisible workhorse of modern life. But Wi-Fi is struggling. And the next phase of the Internet—the Internet of Things—could break it.

    Startups and Internet service providers are developing an application layer that divorces some functions from the base networking standard, IEEE 802.11. Many new features sprouting up within this layer—such as mesh networks (a set of routers that work together to extend wireless coverage) and provisioning tactics (which define how wireless devices connect to networks)—have been developed in response to the Internet of Things.

    It reminds me of the Zigbee mesh-network specification for small, low-power digital radios, which has layers for specific classes of devices built on top of the radio. This has sowed confusion among customers about which Zigbee devices actually work with one another. I fear that Wi-Fi is heading in the same direction.

    Consider a mesh Wi-Fi system. Companies including Eero, Google, and Securifi have created products that promise reliable coverage throughout every square centimeter of a home. But this works only if you buy all of your mesh routers from the same vendor.

    Companies soon realized they could pop a network-based security product into mesh routers and add the ability to prioritize devices. But where does an application layer end and the core Wi-Fi standard begin?

    Diner is working with major Wi-Fi chip vendors to ensure compatibility, and he is open-sourcing some of the necessary code.

    The only solution to fragmentation will be at the Wi-Fi Alliance, which certifies Wi-Fi devices. Brian Bedrosian, vice president of marketing for IoT at Cypress Semiconductor Corp., says simpler provisioning and neighbor-aware networks (which let devices speak to one another without a Wi-Fi access point) must become part of the Wi-Fi certification specification.

    “By comparison, it could be like the Zigbee spec, which has been a disaster,” he says. Hopefully, the Wi-Fi Alliance won’t let that happen.

    Reply
  5. Tomi Engdahl says:

    IEEE 802.3cc standard for 25 Gigabit Ethernet over single-mode fiber published
    http://www.lightwaveonline.com/articles/2018/03/ieee-802-3cc-standard-for-25-gigabit-ethernet-over-single-mode-fiber-published.html?cmpid=enl_lightwave_lightwave_friday_5_2018-03-23&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2045474

    The IEE and the IEEE Standards Association (IEEE-SA) say the recently ratified “IEEE 802.3cc-2017—Standard for Ethernet Amendment: Physical Layer and Management Parameters for Serial 25 Gb/s Ethernet Operation Over Single-Mode Fiber” standard has been published and is now available for use. The specifications point the way toward the development of 25 Gigabit Ethernet transmission technology for enterprise, campus, and metro Ethernet applications at reaches up to 10 and 40 km via single-mode fiber (SMF).

    IEEE 802.3cc defines single-lane 25-Gbps PHYs for the up to 10 km and 40 km use cases.

    Reply
  6. Tomi Engdahl says:

    Shannon Liao / The Verge:
    South Korea fines Facebook ~$370K for slowing user internet connections to its sites by rerouting through Hong Kong and US instead of using local ISPs

    South Korea fines Facebook $369K for slowing user internet connections
    More bad news for Facebook
    https://www.theverge.com/2018/3/21/17147838/facebook-south-korea-slowing-user-internet-connections-reroute

    South Korea’s telecom authority is fining Facebook 396 million won (approximately $369,705) for slowing down user internet connections in 2016 and 2017.

    The Korea Communications Commission (KCC) began investigating Facebook last May and found that the company had illegally limited user access, as reported by ABC News. Local South Korean laws prohibit internet services from rerouting users’ connections to networks in Hong Kong and US instead of local ISPs without notifying those users. In a few cases, such rerouting slowed down users’ connections by as much as 4.5 times. There were 14.5 million Facebook users in South Korea last year and the number is expected to rise to 14.84 million this year, according to Statista.

    Reply
  7. Tomi Engdahl says:

    FCC’s Pai proposes ban on USF use on ‘national security threats’
    http://www.lightwaveonline.com/articles/2018/03/fcc-s-pai-proposes-ban-on-usf-use-on-national-security-threats.html?cmpid=enl_lightwave_lightwave_service_providers_2018-03-26&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2047163

    Federal Communications Commission (FCC) Chairman Ajit Pai said the will have drafted a Notice of Proposed Rulemaking that would bar the use of Universal Service Funds to buy equipment from companies deemed national security threats. The action is in response to the latest wave of concerns on Capitol Hill regarding alleged ties between the Chinese military and intelligence communities and Chinese companies such as Huawei and ZTE.

    Pai will call for a vote on the proposal at a meeting April 17. The announcement comes after reports last Friday that he sent a letter to Congress in which he stated he shares concerns recently expressed about Huawei. For example, in February, Senators Tom Cotton (R-Arkansas) and Marco Rubio (R-Florida) introduced the “Defending U.S. Government Communications Act,” which would prohibit the United States government from buying or leasing telecommunications equipment and/or services from Huawei, ZTE, or any their subsidiaries or affiliates. Congressman Mike Conaway (Texas-11) introduced a similar bill in the House of Representatives in January.

    “Threats to national security posed by certain communications equipment providers are a matter of bipartisan concern,” said Pai in a press statement. “Hidden ‘back doors’ to our networks in routers, switches—and virtually any other type of telecommunications equipment—can provide an avenue for hostile governments to inject viruses, launch denial-of-service attacks, steal data, and more. Although the FCC alone can’t safeguard the integrity of our communications supply chain, we must and will play our part in a government- and industry-wide effort to protect the security of our networks.”

    Reply
  8. Tomi Engdahl says:

    Zayo chosen by Pico for low-latency global backbone network
    http://www.lightwaveonline.com/articles/2018/03/zayo-chosen-by-pico-for-low-latency-global-backbone-network.html?cmpid=enl_lightwave_lightwave_service_providers_2018-03-26&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2047163

    Zayo Group Holdings, Inc. (NYSE: ZAYO) says that financial infrastructure services provider Pico has chosen it for a low-latency global backbone network. The network was designed with two diverse rings to connect Pico’s key financial markets in the U.S., Europe, and Asia, with fully diverse subsea and terrestrial systems, and cable landing stations.

    According to Zayo, the primary ring delivers enhanced low latency, and the secondary ring delivers what Zayo describes as “the next best available latency,” while maintaining 100% physical diversity. The network will leverage Zayo’s infrastructure in North America and Europe, and its global reach relationships with international network providers. Zayo is supporting both global rings for the backbone, and providing diverse connectivity to South Africa for Pico to link the region back to main European financial centers.

    “Zayo’s solution delivers high-performance, low-latency wavelengths across multiple vendor networks,”

    Zayo says this service is part of an ongoing initiative to extend its global reach, enabling worldwide connectivity to over 1,100 data centers with expanded points of presence (PoP) and partnerships. In

    Reply
  9. Tomi Engdahl says:

    NEC supplying 10G-EPON for KDDI’s au Hikari Home 10 Giga FTTH service
    http://www.lightwaveonline.com/articles/2018/03/nec-supplying-10g-epon-for-kddi-s-au-hikari-home-10-giga-ftth-service.html?cmpid=enl_lightwave_lightwave_service_providers_2018-03-26&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2047163

    NEC Corp. (NEC; TSE: 6701) said it will supply a 10G-EPON system to support KDDI Corp.’s au Hikari Home 10 Giga fiber to the home (FTTH) service to enable upstream and downstream speeds of up to 10 Gbps.

    NEC attests that the au Hikari Home 10 Giga high-speed service enables the fastest existing upstream and downstream speeds, referencing an Ovum report stating that “As of 15 January 2018, a 10-Gbps symmetrical residential mass market FTTH service offering at $100 or less would represent the fastest and least expensive offering available.”

    NEC’s 10G-EPON system comprises an optical network unit (ONU) installed within individual subscribers’ homes and an optical line terminal (OLT) installed in KDDI facilities. An OLT is the size of a 4U 19-inch rack, with small, high-density packaging, and supports as many as 8,192 ONUs to minimize the space necessary for installation by telecommunications carriers.

    Reply
  10. Tomi Engdahl says:

    Structured cabling market to reach $11.45B by 2021 at 7.1% CAGR: Report
    http://www.cablinginstall.com/articles/2018/03/transparency-cabling-rpt.html?cmpid=enl_cim_cim_data_center_newsletter_2018-03-26&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2045271

    “Among the different applications of structured cabling, the commercial segment held the largest market share in 2014, holding 35.0% share approximately. The market for structured cabling components used in commercial sector is predicted to expand at a CAGR of 7.1% from 2015 to 2021.

    Increasing demand of Enterprise Resource Planning (ERP) coupled with growing trend of migrating Customer Relationship Management (CRM) to third-party data centers has triggered the need for higher bandwidth in commercial sectors. This in turn is responsible for the huge demand of structured cabling across different commercial sectors which has higher bandwidth and facilitate faster data transmission in comparison to traditional cabling system.

    Reply
  11. Tomi Engdahl says:

    Analyze post-equalization ISI with pulse response
    https://www.edn.com/electronics-blogs/eye-on-standards/4460406/Analyze-post-equalization-ISI-with-pulse-response

    Any signal-integrity engineer will tell you that analyzing closed eye diagrams has never been easy. A transmission channel’s frequency response (or lack of bandwidth) causes inter-symbol interference (ISI), the primary eye-closing culprit. While I’ve written about closed eye analysis techniques before, this time we’ll measure the ISI left over after equalization, the so-called residual ISI. In the process, we’ll see the simple guts of decision feedback equalization (DFE).

    The pulse response is gaining popularity in technology standards that define high-speed serial buses. You can produce a pulse with a pattern generator by transmitting a long string of zeros, a one, and then another long string of zeros, that is, a pulse is a non-return-to zero (NRZ) bit and the pulse response is the same as the SBR (single bit response).

    Reply
  12. Tomi Engdahl says:

    Silicon carbide accelerates the quantum connections

    Researchers from the Moscow Physics and Technology Institute (MIPT) have “rediscovered” the material that can create the basis for very fast quantum connections. According to the study, quantum data transfer can safely increase speed in more than one gigabit per second.

    The biggest expectation of a quantum computer is that it can break the security of classical data transmission networks. But quantum technology also provides a way to neutralize this threat.

    The principle of quantum communication is based on the fact that unknown quantum space can not be copied without changing the original message. That is, the line of quantum communication can not be “listened” without the sender and receiver noticing it. Even a quantum machine would not be of any use to eavesdroppers.

    Photons are the best carriers of quantum bits but can only be used individually. The eavesdropper can capture some of the transmitted photons and get something out of the message.

    The principle of generating a single photon is quite simple, but implementation has proved to be more difficult. MIPT researchers see a solution in silicon carbide, a semiconductor material that has long been forgotten in optoelectronics.

    According to researchers, silicon carbide-based individual photon-emitting diodes can be enhanced to transmit up to several billion photons per second. This can drive quantum technology protocols at data rates of gigabits per second.

    Source: http://www.etn.fi/index.php/13-news/7764-piikarbidi-kiihdyttaa-kvanttiyhteydet

    Reply
  13. Tomi Engdahl says:

    Foxconn Announces Purchase of Belkin, Wemo, and Linksys
    https://hardware.slashdot.org/story/18/03/27/043206/foxconn-announces-purchase-of-belkin-wemo-and-linksys?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Slashdot%2Fslashdot%2Fto+%28%28Title%29Slashdot+%28rdf%29%29

    Foxconn, the Taiwan-based company best-known for manufacturing Apple products announced that one of its subsidiaries (Foxconn Interconnect Technology) is purchasing U.S.-based Belkin for $866 million in cash. “Belkin owns a number of major brands, including Linksys and Wemo,”

    Foxconn announces purchase of Belkin, Wemo, and Linksys
    https://www.androidpolice.com/2018/03/26/foxconn-announces-purchase-belkin-wemo-linksys/

    For years, Taiwan-based Foxconn has been best-known for manufacturing Apple products, but it also builds countless other consumer electronics. Today, one of the company’s subsidiaries (Foxconn Interconnect Technology) announced it is purchasing US-based Belkin for $866 million in cash. Belkin owns a number of major brands, including Linksys and Wemo.

    The buyout would make Foxconn a major player in consumer electronics, instead of just a contract manufacturing company. Belkin primarily sells phone/tablet accessories, but also manufactures networking equipment like routers and Wi-Fi range extenders. The company also sells a range of smart home products under the Wemo brand, like the smart switches we reviewed earlier this year.

    Belkin’s router business is likely to gather scrutiny from government officials, especially since they aren’t particularly secure to begin with. Foxconn pledged to build a $10 billion factory in Wisconsin last year

    Reply
  14. Tomi Engdahl says:

    AT&T, ONF team on multi-gigabit open source PON
    http://www.broadbandtechreport.com/articles/2018/03/at-t-onf-team-on-multi-gigabit-open-source-pon.html?cmpid=enl_btr_weekly_2018-03-27&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24

    AT&T (NYSE:T) and the Open Networking Foundation (ONF) are collaborating to integrate the ONFs’ work on multi-gigabit passive optical networks (PON) with the service automation system, ONAP. ONAP (Open Network Automation Platform) is AT&T’s virtual access project within the Linux Foundation.

    Work will begin to integrate VOLTHA (Virtual Optical Line Termination Hardware Abstraction), the open source software stack powering PON networks, with ONAP. The project is intended to enable agile development progression for virtualized and disaggregated network access for PON networks.

    The work will build upon ongoing field trials of XGS-PON, which is a fixed wavelength symmetrical 10 Gbps PON technology. It also builds upon previous GPON and CORD trials. The trials focused on the viability of disaggregated architecture and the effectiveness of open source and open specs in transforming networking.

    The network uses the following open source software.

    AT&T Open XGS-PON OLT: an OCP Accepted white box OLT
    ONOS: the ONF SDN controller that hosts virtual OLT control applications
    vBNG: a virtual broadband network gateway application to manage subscribers
    VOLTHA: an ONF software project that provides hardware abstraction and an SDN driver for OLT devices

    Reply
  15. Tomi Engdahl says:

    CMOS Optoelectronics: Avalanche-mode silicon LEDs improve optical coupling in CMOS integrated circuits
    https://www.laserfocusworld.com/articles/print/volume-54/issue-01/world-news/cmos-optoelectronics-avalanche-mode-silicon-leds-improve-optical-coupling-in-cmos-integrated-circuits.html?cmpid=enl_lfw_lfw_detectors_and_imaging_newsletter_2018-03-27&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2048233

    In the world of silicon photonics, researchers aim to develop compact, monolithic optical platforms with integrated light sources and photodetectors to perform electronic-to-optical-to-electronic conversion at the speed of light. As with any other optoelectronic integrated circuit (IC), the goal is efficient and high-speed routing and processing of information signals on and off the chip. The same is true for complementary metal-oxide semiconductor (CMOS) platforms—however, the optical interconnect or optocoupler technology has, until now, been limited to bulky infrared (IR) sources that operate at relatively low (kilohertz) speeds.

    To advance optocoupling in CMOS ICs, Ph.D. student Satadal Dutta from the University of Twente (Enschede, Netherlands) has proposed using silicon (Si) light-emitting diodes (LEDs) biased in “avalanche breakdown” mode to generate visible light that is easily detected by standard Si photodiodes in a high-quantum-efficiency process.1 In addition, Dutta and PhD student Vishal Agarwal have modeled and designed a CMOS IC using avalanche-mode LEDs, waveguides, and single-photon avalanche diode (SPAD) detectors that can theoretically operate at gigahertz speeds and single-photon detection sensitivities. Essentially, the all-silicon platform avoids many of the roadblocks presented by hybrid materials systems.

    Reply
  16. Tomi Engdahl says:

    Revenue from data center network equipment hit $13.7 billion in 2017: IHS Markit
    http://www.lightwaveonline.com/articles/2018/03/revenue-from-data-center-network-equipment-hit-13-7-billion-in-2017-ihs-markit.html?cmpid=enl_lightwave_lightwave_datacom_2018-03-27&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2049037

    According to IHS Markit, revenue from data center network equipment totaled $13.7 billion in 2017, an increase of 13% over 2016. The figure includes sales of data center Ethernet switches, application delivery controllers (ADCs), and software-defined enterprise WAN (SD-WAN).

    Physical infrastructure investment continues to accelerate the rise in data center network equipment revenue in the short term. The market research and analysis firm expects that the impact of server virtualization will slow the market in 2018 and 2019, with less (but higher capacity) servers, causing the demand for data center Ethernet switch ports to decline, as well as the move to virtual ADCs.

    “The adoption of lower-priced bare metal switches will cause revenue growth to slow,” said Clifford Grossner, IHS Markit senior research director and advisor, cloud and data center research practice. “The ongoing shift to the cloud not only moves network equipment out of the enterprise data center, but also requires less equipment, as the cloud represents data center consolidation on a wide scale.”

    IHS Markit says that year-over-year, data center network equipment revenue grew in all regions. In 2017, North America and Europe, and Middle-East and Africa (EMEA) each saw a 10% increase. Asia Pacific (APAC) had a 23% rise, and Caribbean and Latin America (CALA) experienced a 2% rise. 25 Gigabit Ethernet (GbE) and 100GbE data center switching ports grew three-fold year-over-year, and developments are underway for new ports of 200GbE and 400GbE. The firm expects shipments to begin in 2019.

    Reply
  17. Tomi Engdahl says:

    The case for fiber deep, digital optics
    http://www.broadbandtechreport.com/articles/2018/03/the-case-for-fiber-deep-digital-optics.html?cmpid=enl_btr_weekly_2018-03-27&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24

    Cable operators seeking to differentiate their video and Internet services from the competition by improving quality of experience (QoE) could be looking toward a “fiber deep” architecture.

    “Older analog equipment does not work very well with new DOCSIS type technologies and DOCSIS Full (Duplex). There are challenges around being able to maintain the plant end as well as provide the (full) amount of bandwidth and capacity,” said Wayne Hickey, advisor, product marketing, Ciena (NYSE:CIEN).

    DOCSIS 3.1 allows for coaxial spectrum to be used more efficiently with RF signals bonded together. The actives in the plant cause a challenge however, as does the fact that analog fiber is limited in its capacity. Full Duplex DOCSIS will require even more bandwidth.

    In a distributed access architecture (DAA), one analog fiber node (usually serving more than 500 customers) is replaced with 10-12 digital fiber nodes (or R-PHY devices). The service groups are reduced to about 64 customers. Converting more of the access network from analog to digital increases bandwidth per home passed. Connectivity improves as does quality of service.

    An additional benefit is eliminating amplifiers, and by bringing the optical-to-electrical conversion closer to subscribers, power and maintenance are decreased.

    “What it means is less devices,” Hickey said. “If you have to do an analog sweep twice a year, there is a cost burden associated with that …. By pushing the optical fiber closer to the home, the network is more adaptive and responsive to customers’ needs. When they need or want things, it can be controlled (through software) rather than rolling trucks.”

    Where data centers are concerned, fiber deep means virtualizing services and providing more localized content, which also means better quality of service, particularly for time sensitive content.

    Reply
  18. Tomi Engdahl says:

    Cisco separates switching and routing software from hardware
    https://www.theregister.co.uk/2018/03/28/cisco_disaggregation_strategy/

    Open networking finally comes to Switchzilla as IOS XR, IOS XE, Nexus OS added to disaggregation strategy

    Cisco has taken a long, hard look at the biggest bullet in its kit, and bitten down: the company has announced it will separate its router and switch software from the hardware that hosts it.

    The “kill-your-darlings” decision is in response to demand from hyperscale Web outfits and service providers, rising sales of white-box OEMs, and growing competition from software-only “niche vendors”, Switchzilla said in its announcement.

    The decision marks a huge reversal on former CEO John Chambers’s view of the world viewed the world. Early in the days of disaggregation and white-box offerings, Chambers stuck to his guns, saying customers wanted the company to respond to their business needs in ways that “point product” vendors could not.

    Switchzilla first dipped its toe in the new software-defined world with its Enterprise Network Compute System, which early last year became a fully-fledged NFV (network function virtualisation) solution. At that time, Cisco also unveiled its first virtual customer premises equipment product.

    Now, Cisco is adding:

    A software-only IOS XR product for service providers;
    Enhancements to its IOS XE; and
    The Nexus OS also split from its hardware.

    Cisco said the disaggregated IOS XR allows the software to run on Cisco or merchant silicon-based switches and routers, or as a virtualised instance on x86-based clouds. It’s also available on “specific” third party devices

    In the data centre

    For data centre customers, as the company’s Roland Acra wrote, there’s a “cloud scale” SAI (switch abstraction interface) that lets the switch run other vendor’s operating systems.

    Acra says Microsoft has already bolted its SoNIC network operating system onto its Nexus 9200 and 9300 platforms, and quotes Microsoft Azure Networking CVP Yousef Khalidi as saying it means Redmond and cloud operators can use “the same software stack across a variety of switch hardware platforms”.

    Reply
  19. Tomi Engdahl says:

    Photon/Phonon Conversion Sheds New Light on Optical Signal Processing
    http://www.electronicdesign.com/analog/photonphonon-conversion-sheds-new-light-optical-signal-processing?NL=ED-003&Issue=ED-003_20180328_ED-003_760&sfvc4enews=42&cl=article_2&utm_rid=CPG05000002750211&utm_campaign=16230&utm_medium=email&elq2=7fa21416fc32430e90dd63051fa4e5a5

    The processing of optical signals, with their real-time, streaming nature, would be greatly enhanced if there was a mechanism for writing, storing, and reading back the data. It’s the same situation with our familiar electronic data, of course, which is why “memory” is a major system function. Phonons—quanta of coherent acoustic vibrations—may offer a way to store and retrieve optical signals, and at a speed commensurate with the optical domain

    To facilitate processing of signals in the high-speed optical domain, there must be a way to store and retrieve. Converting them to acoustic phonons may offer a solution to this vexing challenge.

    To pursue this objective, researchers at the University of Sydney in Australia have demonstrated a buffer built as an optical waveguide that works by coherently translating the optical data to an acoustic hypersound wave. Optical information is extracted using a complementary process. The result is the storage of phase and amplitude of optical information with gigahertz bandwidth as well as multichannel operation at independent wavelengths, and with negligible crosstalk.

    Their detailed paper in Nature, “A chip-integrated coherent photonic-phononic memory,” details the theory, actual implementation, and results. The objective is to develop components for storing or delaying optical signals, enabling mostly optical processing and thus the speed and performance benefits that development would bring. Signal “delay” is needed so that data can be briefly stored and managed inside the integrated device for processing prior to retrieval and transmission.

    Reply
  20. Tomi Engdahl says:

    AAEON FWS-7360 10GbE Network Platform
    https://www.eeweb.com/profile/eeweb/news/aaeon-fws-7360-10gbe-network-platform

    AAEON announces the launch of the 1U enterprise-grade FWS-7360. This high-performance network appliance has a range of features including 10GbE SFP+ fiber ports to ensure that businesses have constant access to secure, high-speed connections.

    Powered by Intel® Atom™ C3758, C3858, or C3958 processors (formerly Denverton), the FWS-7360 offers outstanding levels of performance and the guarantee of reliable, high-speed data compression and encryption that comes with Intel QuickAssist Technology.

    The processors’ eight- to sixteen-core configurations mean you can count on higher data transmission throughput rates and a reduced risk of system bottlenecks and data loss. The high number of cores and the appliance’s WiFi and 4G LTE capabilities also play a vital role in its ability to deliver SDN, NFV, and vCPE applications.

    With two 10GbE SFP+ fiber ports complementing five copper Ethernet ports and two pairs of LAN bypass support, the FWS-7360 provides a full range of high-speed, conventional, and protected network connectivity. AAEON’s latest 1U network appliance also stands out from the competition in that its 10GbE ports are connected directly to the CPU instead of an expansion slot.

    “The built-in fiber ports are a very strong feature, but almost as important is the appliance’s flexibility,”

    Larger Workloads Handled Faster: the AAEON FWS-7360 10GbE Network Platform
    http://www.aaeon.com/en/ni/press-release-fws-7360

    Reply
  21. Tomi Engdahl says:

    FCC approves SpaceX plan for 4,425-satellite broadband network
    https://techcrunch.com/2018/03/29/fcc-approves-spacex-plan-for-4425-satellite-broadband-network/?utm_source=tcfbpage&utm_medium=feed&utm_campaign=Feed%3A+Techcrunch+%28TechCrunch%29&utm_content=FaceBook&sr_share=facebook

    SpaceX has a green light from the FCC to launch a network of thousands of satellites blanketing the globe with broadband. And you won’t have too long to wait — on a cosmic scale, anyway. Part of the agreement is that SpaceX launch half of its proposed satellites within six years.

    The proposed service, which will be called Starlink, was opposed by several existing satellite operators like OneWeb and Spire. They’re rightly concerned that another operator in space — especially one that wants to launch thousands of satellites — will crowd both spectrum and orbit.

    Reply
  22. Tomi Engdahl says:

    Bloomberg:
    How Google, carriers, and startups plan to capitalize on the Citizens Broadband Radio Service spectrum, likely to be freed from US military control this year

    Carriers Are Hoarding America’s Bandwidth. Google Just Wants Them to Share
    https://www.bloomberg.com/news/articles/2018-03-29/google-led-plan-to-upend-wireless-industry-gains-momentum

    A bold plan to open up spectrum could rock the wireless industry and prove a $260 billion win for consumers. Guess-who would also make out all right.

    Reply
  23. Tomi Engdahl says:

    Wall Street Journal:
    Rural internet and cable providers worry that regulations against Huawei will hurt their margins by forcing them to buy pricier equipment from Western suppliers — Some rural internet providers rely on telecom gear from China’s Huawei, which faces potential new restrictions from FCC, Congress

    In U.S. Brawl With Huawei, Rural Cable Firms Are an Unlikely Loser
    https://www.wsj.com/articles/caught-between-two-superpowers-the-small-town-cable-guy-1522152000

    Some rural internet providers rely on telecom gear from China’s Huawei, which faces potential new restrictions from FCC, Congress

    Here is a potential casualty of the U.S. government’s escalating fight against Huawei Technologies Co.: rural phone companies and internet providers that depend on the Chinese giant’s gear to connect their customers.

    Large wireless providers including AT&T Inc. have long steered clear of Huawei, which has been effectively barred from big U.S. business since a 2012 congressional report alleged the Chinese government could force the company to exploit knowledge of how its equipment is designed to spy or launch cyberattacks—a charge Huawei has denied.

    But many regional American providers of wireless, TV and internet services have flocked to Huawei, attracted by what they say are Huawei’s cheaper prices, quality products and attentive customer service.

    On Monday, the Federal Communications Commission proposed making it harder for these smaller carriers to pay for future purchases of telecom equipment from Huawei and Chinese peers.

    Meanwhile, a congressional bill with some bipartisan support aims to prohibit carriers with any substantial amount of installed Chinese telecom equipment from federal-government contracts.

    Huawei, the world’s top maker of cellular-tower electronics and a major manufacturer of equipment for cable and internet providers, has been actively courting small-town internet companies that wanted to replace old-fashioned landlines with high-speed internet connections—no small feat in a country where most rural residents are stuck with dial-up speeds.

    Many of these customers now worry the new heat over Huawei in Washington may rob them of what has so far been an important alternative to Western suppliers. Others worry that if Huawei exits the U.S. completely, it will leave them without the customer and technical support they need to maintain the Huawei hardware they already own.

    Huawei’s standoff with the U.S. government has been a boon to Sweden’s Ericsson AB and Finland’s Nokia Corp. , which dominate the $30-billion-a-year market for wireless equipment in the U.S. It also shields domestic companies like Silicon Valley’s Cisco Systems Inc., which make electronics such as routers for cable and internet providers.

    Reply
  24. Tomi Engdahl says:

    SYN-SYN-ACK – Your networking news roundup has arrived!
    Aruba’s AI moves, Marvell and Nvidia vehicle tech, and yes, the Open Networking Summit
    https://www.theregister.co.uk/2018/03/29/network_roundup_march_29/

    Network admins: unpatched MikroTik routers are being scanned by a botnet again.

    Netlab360 is one of the organisations to identify increasing scans on Port 8291, and says it’s associated with a new Hajime malware attack.

    Aruba adds AI

    Aruba has launched an analytics offering called NetInsight. It’s 2018 so it offers AI network performance analytics and assurance solution.

    Part of Aruba’s Mobile First Architecture, NetInsight is an anomaly-watcher that provides response and optimisation suggestions “based on data that is specific to user connectivity and RF performance attributes”.

    Reply
  25. Tomi Engdahl says:

    Global PLC splitter consumption reaches nearly 33 million units: ElectroniCast
    http://www.lightwaveonline.com/articles/2018/03/global-plc-splitter-consumption-reaches-nearly-33-million-units-electronicast.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-03-29&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2052151

    In 2017, the worldwide consumption volume of PLC component-level (bare-fiber) splitters reached approximately 32.796 million units, according to the market review and forecast report.

    PLC splitters distribute optical signals from an input optical fiber port to multiple optical fiber output ports. ElectroniCast anticipates that by enabling a single passive optical network (PON) interface to be shared among multiple subscribers, PLC component splitters will remain a key contributing factor in deploying optical fiber closer to the subscriber via fiber-to-the-home (FTTH), fiber-to-the-building (FTTB), and fiber-to-the-premises (FTTP).

    PLC splitters are available in several configurations,

    PON, FTTx, and other telecom network applications continue to be the dominant markets for PLC splitters

    In 2017, PLC splitter compact devices used in FTTx, PONs, and other telecom (and CATV) service provider network applications dominated in global consumption volume, accounting for 97.2% relative market share, or 31.878 million units, reports ElectroniCast.

    Reply
  26. Tomi Engdahl says:

    Inphi shipping LightSpeed-III M200 100G/200G coherent DSP
    http://www.lightwaveonline.com/articles/2018/03/inphi-shipping-lightspeed-iii-m200-100g-200g-coherent-dsp.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-03-29&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2052151

    Inphi Corp. (NYSE: IPHI) said it has begun shipping the production version of its M200 LightSpeed-III coherent digital signal processor (DSP). The device features what the company asserts is ultra-low power and high performance to support 100G and 200G data rates for long-haul, metro, and data center interconnect (DCI) applications. Several Tier 1 OEMs have committed to commercial deployments of the M200 offering in the second half of 2018, says Inphi.

    The M200 is the first offering in Inphi’s 16-nm LightSpeed-III SoC device family, using Inphi’s DSP and FEC technology, and its internal analog expertise and SerDes IP. The M200 DSP supports two host SerDes with selectable 10-Gbps and 28-Gbps NRZ interfaces with 100G client FEC termination.

    Additionally, the M200 enables OSNR performance of 17.5 dB for 200G 16QAM, which is crucial to enable the large-scale transition from 100G to 200G in metro networks, Inphi attests. To provide this performance while enabling spectral efficiency that is 30% higher than 8QAM-based competitors, M200 uses 16QAM modulation.

    According to Inphi, the M200 enables 200G deployments on legacy and current networks that have been deployed with 50-GHz grid spacing, and maximizes the number of ROADM nodes covered, while eliminating the need for electrical regeneration.

    Reply
  27. Tomi Engdahl says:

    Menara Networks sampling digital coherent CFP2-DCO transceivers to enable data center interconnection
    http://www.lightwaveonline.com/articles/2018/03/menara-networks-sampling-digital-coherent-cfp2-dco-transceivers-to-enable-data-center-interconnection.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-03-29&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2052151

    Menara Networks said it has begun sampling to select customers its digital coherent CFP2-DCO transceivers to enable ZR, data center interconnect (DCI), metro, and long-haul applications with maximum versatility.

    Menara’s digital CFP2-DCO “system-in-a-module” is based on a 16-nm coherent processor, and offers full DWDM coherent transmission for 100G and 200G per wavelength in a CFP2 MSA small form factor. It performs all required adaptive PM-QPSK and PM-QAM 16 modulation, digital signal processing for linear and non-linear fiber impairments compensation, G.709 compliant OTN framing, and various hard and LDPC-based soft decision forward error correction encoding/decoding. Menara’s coherent CFP2-DCO is compatible with 100GbE, CAUI, OTL4.10, and OTL 4.4 interfaces, is full C-band tunable, and supports ITU-T G.694.1 6.25-GHz flexible grid.

    Reply
  28. Tomi Engdahl says:

    Ciena taps open source for network software
    http://www.broadbandtechreport.com/articles/2018/03/ciena-taps-open-source-for-network-software.html?cmpid=enl_btr_weekly_2018-03-29&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24

    Ciena (NYSE:CIEN
    ) is introducing upgraded policy capabilities into its Blue Planet automation platform that incorporates the architectural framework from the Linux Foundation’s open-source Open Networking Automation Platform (ONAP) to aid network providers’ transition to software-centric networks. The new Blue Planet policy subsystem is intended to support closed-loop automation use cases, such as preventing a potential failure or dynamically scaling capacity to support growing bandwidth demands. Additionally, it is intended to give operators more control to create intent-based policies to better guide their networks to address potential obstacles and meet growth objectives.

    Ciena Blue Planet is adding support for ONAP virtual network function (VNF) descriptors and packaging specifications to the already supported ETSI ISG NFV Release 2 specification. Blue Planet’s support for ONAP VNF descriptors and packaging specifications are available today, while the enhanced policy subsystem is expected to be available in the second half of 2018.

    Ciena CTO Steve Alexander said: “The future of networking lies with openness and the concept of ‘choice,’ allowing network operators to select best-of-breed technologies and solutions.

    Reply
  29. Tomi Engdahl says:

    5 commonly asked questions about fiber-optic cabling
    http://www.cablinginstall.com/articles/2018/03/5-questions-about-fiber-commscope-mike-cooper.html?cmpid=enl_cim_cim_data_center_newsletter_2018-04-02&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2053821

    1. How long of a span can I suspend self-supporting fiber?

    CommScope provides span lengths in three categories: NESC vehicular access, NESC pedestrian access, and infinity within each NESC loading category: heavy, medium and light. When a self-supporting fiber cable is latched to a support strand such as a ¼-inch 6.6M EHS strand, the self-support span limitations no longer apply because the load is being placed on the strand and not the cable.

    The only sure way to know the limitations is to review the specifications for the cable.

    2. Can I bend fiber around a sharp corner?

    In the past, we were limited to where and how we placed slack storage.
    There are still fiber-optic glass bend radius standards. ITU-T G.657.A1 has a minimum bend radius of 10mm, G.657.A2/B2 at 7.5mm and G.657.B3 down to 5mm. With specified bend radii, you need to maintain clean systems with the proper RBR. There may still be bend loss, but maybe not in the same way as the past. Historically, you would trace the fiber to physically see the bend. With today’s fiber, the bend loss might indicate more of an improper seating of a connector or a routing issue in a splice tray.

    3. I have heard about 8-fiber connectivity, but use 12-fiber MPOs. Do I need to switch?

    There is a saying about how “less is more,” but in this case, the opposite is true. The short answer is no. You do not need to switch. If managed well, a 12-fiber MPO infrastructure with more fibers per connector provides 50 percent more usable fibers per connector in the same footprint. This is valuable as evolving standards continue to use duplex fibers as connectivity options through at least 100 Gbits/sec.

    The main reason for introduction of 8-fiber MPO connectors was to provide application support of parallel signals using 8 of the available 12 fiber positions of the industry-standard 12-fiber MPO connector. This application typically occurs in a QSFP (Quad Small Form-factor Pluggable) transceiver.

    4. Will fiber be the best solution to connect cellular network radios in the future?

    That is the consensus. Mobile network operators will opt for fiber as the preferred technology for backhaul and fronthaul to cellular network radios wherever possible because of the ever-increasing bandwidth requirements of today and into the future.

    5. How does one confirm compliance of fiber-optic cables to new European CPR regulations for in-building applications?

    Introduced in 1989 by European regulators, the Construction Product Directive (CPD), later the Construction Product Regulation (CPR), was designed to ensure materials and equipment that make up buildings such as offices, schools, and shops are safe from fire hazards and other risks.

    All manufacturing facilities serving the European market will need to have been audited and approved by Notified Bodies, and the building industry will be required to work only with vendors that have carried out extensive testing with these Notified Bodies, and can provide the proper Declarations of Performance for their cabling products, organized in Euroclasses A to F.

    Every product will be required to carry the CE label appropriate to the applicable Euroclassification. The type of testing is determined by the type of verification system being used by the Notified Body.

    Reply
  30. Tomi Engdahl says:

    Ericsson says it owns 50% of the 5G wireless networks market
    http://www.cablinginstall.com/articles/pt/2018/03/ericsson-says-it-owns-50-of-the-5g-wireless-networks-market.html?cmpid=enl_cim_cim_data_center_newsletter_2018-04-02&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2053821

    Ericsson’s CEO said the company has won roughly 50% of the world’s 5G contracts so far. He also said that Ericsson has reversed its declines in the global wireless network business overall, and during the second half of 2017 the company was able to gain share in the sector.

    Ericsson: We’re gaining share in wireless networks, own 50% of 5G market
    https://www.fiercewireless.com/tech/ericsson-we-re-gaining-share-wireless-networks-own-50-5g-market

    Ericsson’s CEO said the company has won roughly 50% of the world’s 5G contracts so far. He also said that Ericsson has reversed its declines in the global wireless network business overall, and during the second half of 2017 the company was able to gain share in the sector.

    “We have been working hard to turn around the development,” President and CEO Börje Ekholm said in a release from the company. “It is therefore satisfying that we, after several years of decreasing market share, have started to increase our share, and doing so with improved gross margin.”

    Added Ekholm: “2017 was a tough year with a continued declining market. We are far from satisfied with our performance and have taken a number of actions to turn around the development and improve profitability, to build a strong Ericsson for the long term.”

    But Ericsson’s statements on 5G don’t appear to line up exactly with those from some analysts. Analysts at Raymond James recently said that it’s too early to call winners, but clearly Ericsson, Huawei and Nokia dominate the wireless infrastructure market. Specifically, the Wall Street firm said Nokia has announced 50 5G trials, and Ericsson has signed 38 operator agreements.

    Reply
  31. Tomi Engdahl says:

    ONF targets Stratum open source reference platform at white box switch support
    http://www.lightwaveonline.com/articles/2018/03/onf-targets-stratum-open-source-reference-platform-at-white-box-switch-support.html?cmpid=enl_lightwave_lightwave_friday_5_2018-03-30&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2052612

    The Open Networking Foundation (ONF) has unveiled Stratum, an open source project that aims to promote the development of white box switches through the implement a software platform that leverages the ONF’s newly announced next-generation software-defined networking (SDN) interfaces. The initiative has attracted the participation of companies from across the data center networking community, including both telecom services and cloud providers. The latter demographic includes Google, which will help get the project rolling by providing what will amount to the first revision of source code for Stratum.

    Despite the efforts of previous initiatives, current open system white box switches haven’t fully achieved plug-and-play status, according to Timon Sloane, vice president of marketing and ecosystem at the ONF. The Stratum project will seek to address this shortfall by leveraging the newly released SDN interfaces, which improve lifecycle management and control. Sloane described these interfaces as successors to OpenFlow that enable improved negotiation between the network operating system (NOS) and the data plane. They more fully enable a pipeline “contract” as well as pipeline control, configuration, and operations. The result is more deterministic performance, he said.

    The ONF currently envisions four use cases for Stratum software:

    Cloud SDN Data Plane: This is the use case in which Google has the most immediate interest. As noted, Google has donated an initial code base for Stratum from its production codebase in the expectation that it will be able to use Stratum in its production networks quickly. Google will pair Stratum with the company’s internal SDN controller to manage and control Stratum-based white box switches. Other network operators with proprietary controllers could follow Google’s lead.
    Cloud SDN Fabric Platform: Those who use the ONF’s ONOS plus Trellis software can run it natively over Stratum as a P4 data plane switching layer. The result is an open source approach to programmable SDN data center spine-leaf fabrics, the ONF says.
    Central Office Re-architected as a Datacenter (CORD): CORD also runs over Trellis and ONOS, which means Stratum can play a role in CORD environments as well.
    Thick Switches with Embedded Control: While Stratum will be optimized to support “thin switch” implementations, developers also can use it with an embedded NOS running within the same switch. This “thick switch” configuration enables the replication of the traditional model of embedded management and control (e.g., Netconf, SNMP, BGP) within a single system. As a result, more conventional, tightly integrated systems can leverage Stratum to benefit from the latest SDN interfaces.

    Reply
  32. Tomi Engdahl says:

    Revenues from data center network equipment hit $13.7 billion in 2017: IHS Markit
    http://www.lightwaveonline.com/articles/2018/03/revenue-from-data-center-network-equipment-hit-13-7-billion-in-2017-ihs-markit.html?cmpid=enl_lightwave_lightwave_friday_5_2018-03-30&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2052612

    According to IHS Markit, revenue from data center network equipment totaled $13.7 billion in 2017, an increase of 13% over 2016. The figure includes sales of data center Ethernet switches, application delivery controllers (ADCs), and software-defined enterprise WAN (SD-WAN).

    Physical infrastructure investment continues to accelerate the rise in data center network equipment revenue in the short term. The market research and analysis firm expects that the impact of server virtualization will slow the market in 2018 and 2019, with less (but higher capacity) servers, causing the demand for data center Ethernet switch ports to decline, as well as the move to virtual ADCs.

    “The adoption of lower-priced bare metal switches will cause revenue growth to slow,” said Clifford Grossner, IHS Markit senior research director and advisor, cloud and data center research practice. “The ongoing shift to the cloud not only moves network equipment out of the enterprise data center, but also requires less equipment, as the cloud represents data center consolidation on a wide scale.”

    IHS Markit says that year-over-year, data center network equipment revenue grew in all regions. In 2017, North America and Europe, and Middle-East and Africa (EMEA) each saw a 10% increase. Asia Pacific (APAC) had a 23% rise, and Caribbean and Latin America (CALA) experienced a 2% rise. 25 Gigabit Ethernet (GbE) and 100GbE data center switching ports grew three-fold year-over-year, and developments are underway for new ports of 200GbE and 400GbE. The firm expects shipments to begin in 2019.

    For the full-year 2017, SD-WAN market revenue reached $444.1 million, and revenue is expected to total $3.6 billion by 2022, reports IHS Markit. Data center Ethernet switch revenue hit $11.4 billion in 2017, representing a 13% increase over the previous year. The firm says that bare metal switch revenue was up 60% year-over-year in the fourth quarter of 2017, while ADC revenue was down 5% year-over-year in 2017.

    Reply
  33. Tomi Engdahl says:

    Using Extreme Analytics To Turn Telco Networks Into Customer Insight Channels In The Post-Big Data Era
    https://www.forbes.com/sites/kinetica/2018/03/13/using-extreme-analytics-to-turn-telco-networks-into-customer-insight-channels-in-the-post-big-data-era/#18a7b1b059b6

    Beyond the satellite networks, the telecommunications industry is taking fire from all angles. For instance, Google and Facebook are set to attract 84 percent of global spending in digital advertising, raising fears of a digital duopoly.

    Despite the new competitive landscape, telco operators are facing a massive opportunity for growth. As more users, devices, and things get online, mobile networks are the common thread connecting these digital relationships. Every enterprise in the world is struggling to understand their customer relationships across the digital and physical worlds, and mobile is the fabric that ties them together. In addition, emerging technologies such as 5G provide a new digital highway for transmitting more data than ever before. This technology can capture incredible volumes of information about customer interactions. In short, data is the new currency for telco operators.

    The problem is, legacy big data infrastructure solutions were not designed to handle the unpredictability of streaming data sources, nor were they designed to utilize data while fresh. Now, telco operators must mine real-time data and instantly analyze it

    We have entered the post-big data era. In this new world, telcos need to translate massive volumes of complex data into digital insight at unparalleled speed, with streaming data analysis, visual foresight, and streamlined machine learning. Without these capabilities, it becomes impossible to maintain business in motion.

    Reply
  34. Tomi Engdahl says:

    Molex Showcases Next-Generation Optical Solutions at OFC 2018
    Explore high-speed, high-density solutions at the forefront of high-performing networks
    https://www.molex.com/molex/news/display_news.jsp?channel=New&channelId=-8&oid=2343&pageTitle=Molex+Showcases+Next-Generation+Optical+Solutions+at+OFC+2018

    Molex Debuts 100G PAM-4 Based 25G/50G/100G/400G Product Solutions
    Next Generation Data Center and 5G Wireless Applications
    https://www.molex.com/molex/news/display_news.jsp?channel=New&channelId=-8&oid=2344&pageTitle=Molex+Debuts+100G+PAM-4+Based+25G%2F50G%2F100G%2F400G+Product+Solutions+

    Molex today announced a portfolio of 100G and 400G products based on its 100G PAM-4 optical platform, including multi-Rate 25G/50G/100G PAM-4 DWDM QSFP28, 100G FR QSFP28, 400G DR4 and 400G FR4 QSFP-DD and OSFP.

    100G PAM-4 platform-based product solutions include:

    • 25G/50G/100G Multi-Rate PAM4 DWDM QSFP28:
    o Supports 40×100 Gbps DWDM with a total of 4 Tbps per fiber pair
    o Flexible data rate programmability to support 25 Gbps and 50 Gbps applications
    • 100G FR QSFP28:
    o Compliant with 100 Gbps Lambda MSA
    o Supporting 100G connectivity up to 2km
    • 400G DR4 and 4x100G FR QSFP-DD/OSFP:
    o Compliant with 100G Lambda MSA
    o Supporting 400 Gbps connectivity up to 500m
    o 4X100G FR supporting four times of 100G FR up to 2km
    • 400G FR4 QSFP-DD/OSFP:
    o Compliant with 100G Lambda MSA
    o Supporting 400 Gbps connectivity up to 2km

    Reply
  35. Tomi Engdahl says:

    Cloudflare touts privacy-friendly 1.1.1.1 public DNS service. Hmm, let’s take a closer look at that
    We’ll share query data, but only with these really trustworthy researchers
    https://www.theregister.co.uk/2018/04/03/cloudflare_dns_privacy/

    Cloudflare has revealed a deal with regional internet registry APNIC to provide a possibly more privacy-conscious DNS resolver at a prestige network address, 1.1.1.1.

    The biz contends DNS – which translates human-friendly domain names like theregister.com into numeric IP addresses, such as 159.100.131.165, used by software – lacks privacy protection. That largely undisputed claim has become more noteworthy since the US Congress last year dropped rules that prohibited ISPs from selling users’ browsing data.

    “Your ISP, and anyone else listening in on the internet, can see every site you visit and every app you use – even if their content is encrypted,” the company says on its 1.1.1.1 website. “Creepily, some DNS providers sell data about your Internet activity or use it target you with ads.”

    Cloudflare’s 1.1.1.1 isn’t primarily a website; it’s a DNS lookup service that, when queried by browsers and other software, asks around to various servers where to find the authoritative name server to resolve a particular domain to a network IP address.

    Ironically for a project predicated on privacy, Cloudflare is sharing DNS query data with APNIC Labs, a part of Asian registry APNIC, in exchange for the use of its 1.1.1.1 network address.

    The research relationship is set to run for at least five years, after which it may be renewed and APNIC will consider permanently allocating the 1.1.1.1 IP address – along with 1.0.0.1 – to Cloudflare.

    Logging

    In this Cloudflare’s venture is similar to Google’s Public DNS (8.8.8.8), which claims that it keeps some data for just 24 to 48 hours. Google, however, keeps other non-personally identifiable information for longer periods.

    Sure enough, Cloudflare has positioned its DNS service as an alternative to Google’s.

    Reply
  36. Tomi Engdahl says:

    Software-defined Global Network as a Service Firm Meta Networks Emerges From Stealth
    https://www.securityweek.com/software-defined-global-network-service-firm-meta-networks-emerges-stealth

    Meta NaaS Provides a Software-defined Virtual ‘Overlay’ to Existing Disjointed Physical Networks

    Emerging from stealth with $10 million in seed funding led by Vertex Ventures and the BRM Group, Tel Aviv-based Meta Networks has launched Meta NaaS — a secure software-defined virtual private network aimed at redefining the concept of distributed, cloud-employing corporate networks.

    The advent of public and private cloud services and offerings, together with the growth of mobile computing and remote working, plus the tendency for most companies to combine all of these with their own on-premise resources has had one major and well-recognized effect: there is no longer a physical network perimeter that can be defined and protected. Solutions generally require point products for every device, aimed at protecting the device and its communication to other parts of the network. This rapidly becomes very complex with multiple points of possible failure.

    Meta NaaS provides a software-defined virtual ‘overlay’ to existing disjointed physical networks. It is user-centric, draws on the principle of zero-trust, and brings together all aspects of remote users, mobile devices, separate branch offices, on premise data centers and cloud apps within one single software-defined overlay. It creates a new perimeter in the cloud.

    Like Google’s BeyondCorp, the user is key. Every user device is given a unique permanent identity at the packet level, but is also given access to an always-on virtual private network (VPN). A global distribution of PoPs ensures high performance in accessing and using the VPN from any location, and all corporate traffic from corporate users is securely sent to the NaaS before being delivered to its destination. This includes both internal resources and internet traffic — and security is handled in the NaaS rather than at the device.

    Reply
  37. Tomi Engdahl says:

    Comcast: DOCSIS 3.1 shines in the MDU market
    http://www.broadbandtechreport.com/articles/2018/04/comcast-docsis-3-1-shines-in-the-mdu-market.html?cmpid=enl_btr_weekly_2018-04-03&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24

    In the competitive multiple dwelling unit (MDU) market, amenities matter to existing and potential residents. What kind of high-speed Internet can they access? How convenient is moving in? Are there options when it comes to communicating with property managers?

    What doesn’t necessarily matter is how these services are delivered, said Mike Slovin, VP of Comcast’s (NASDAQ:CMCSA) Xfinity Communities, noting that while fiber is great, particularly in greenfield properties, retrofitting existing buildings is costly and complicated.

    “Residents don’t really care about the physical network infrastructure; they care about speed,” Slovin said.

    DOCSIS 3.1 can deliver the same gigabit speeds as fiber, and all residents have to do is swap out a modem. No cracking of walls or pulling of fiber necessary.

    “Best of all, DOCSIS 3.1 is infrastructure agnostic, meaning it can run over legacy coaxial cable network or fiber,” Slovin said. “(Also,) while fiber may seem like the new kid on the block, DOCSIS 3.1 is still young and improving. It’s been updated several times since CableLabs first released it in late 2013, and I’d argue that DOCSIS 3.1 is the true forward-facing technology based not only on its ability to deliver top speeds today, but to continue growing its capabilities into the future.”

    Reply
  38. Tomi Engdahl says:

    Revenue from optical DCI hardware reached $2.6 billion in 2017: IHS Markit
    http://www.lightwaveonline.com/articles/2018/04/revenue-from-optical-dci-hardware-reached-2-6-billion-in-2017-ihs-markit.html?cmpid=enl_lightwave_lightwave_datacom_2018-04-03&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2056300

    According to a biannual IHS Markit report, optical data center interconnect (DCI) hardware revenue increased by 26% worldwide year-over-year, reaching $2.6 billion in 2017. IHS Markit predicts that this market will surpass $5 billion in annual sales by 2022, representing nearly 30% of all wavelength division multiplexing (WDM) equipment spending.

    Reply
  39. Tomi Engdahl says:

    Price decline in 100GbE device shipments impacts vendors profitability in 2017: LightCounting
    http://www.lightwaveonline.com/articles/2018/04/price-decline-in-100gbe-device-shipments-impacts-vendors-profitability-in-2017-lightcounting.html?cmpid=enl_lightwave_lightwave_datacom_2018-04-03&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2056300

    100GbE device shipments reached close to 2.9 million units in 2017 and is expected to surpass 5 million in 2018. In 2017, prices declined faster than anticipates, resulting in profitability concerns among vendors. Since buyers had placed duplicate orders, this eliminated some of the 100GbE demand, reports LightCounting. Much investment is needed to develop 400GbE products, and the market research firm does not anticipate that customers will purchase these products until ideal pricing is available.

    Reply
  40. Tomi Engdahl says:

    FTBx-88400NGE Power Blazer – 400G multiservice test solution for the field
    https://www.exfo.com/en/products/ftbx-88400nge-field/

    Network infrastructure planners must deal with skyrocketing demands for more bandwidth, including in the data center interconnect (DCI) or even in core networks, as service providers are constantly expanding their networks while seeking efficient and cost-effective ways to deploy high-speed links

    Reply
  41. Tomi Engdahl says:

    The growth in the revenue of the operators is halted

    5G comes and IoT comes, so operators should be facing cat days. Well it is not. The Strategy Analytics research predicts that operators’ revenue will slowly increase until 2021. Then growth stops.

    According to the research institute, operators will invoice the accession of some $ 881 billion in 2021, which is about EUR 715 billion in euros. This is only 3 percent more than the operators’ revenue this year.

    According to Strategy Analytics, revenues will not increase even though the number of subscribers for the operators is developing as expected. There are currently 7.7 billion cell phones / mobile subscriptions in the world. In 2023 the number has increased to nine billion.

    Although revenue from access does not significantly increase, operators need to invest a lot of money in updating their networks. In the next few years, most money will be spent on the purchase of 5G base stations.

    According to Strategy Analytics, next year, 5G networks will have five million users. In 2023 the number has increased to 577 million, or 5G, in many markets has become quite common technology.

    Source: http://www.etn.fi/index.php/13-news/7794-operaattorien-tulojen-kasvu-pysahtyy

    Reply
  42. Tomi Engdahl says:

    Indented Arrays Boost Full-Duplex Systems
    http://www.mwrf.com/components/indented-arrays-boost-full-duplex-systems?NL=MWRF-001&Issue=MWRF-001_20180403_MWRF-001_159&sfvc4enews=42&cl=article_2_b&utm_rid=CPG05000002750211&utm_campaign=16378&utm_medium=email&elq2=e326e0f2a820412eaf74ac722e8de849

    The use of indented antenna arrays can improve the isolation between transmit and receive functions in full-duplex communications systems.

    The growing use of simultaneous transmission and reception (STAR) in full-duplex communications is literally putting the heat on some of the antennas in these systems, since signal leakage from the transmit antennas can double as interference for the receive antennas. Suppression or rejection of such transmit-antenna leakage can be challenging, leading some designers to explore the use of indented antenna arrays in full-duplex communications systems. The basic idea is to break the symmetry between the transmit and receive signal paths by indenting the antenna elements, thus creating an antenna aperture that is curved and less likely to cause self-interference.

    The use of STAR signals within the same frequency band has often been the operating approach of continuous-wave (CW) radar systems and it is increasingly being used in full-duplex communications systems. The approach eliminates the need to separate the frequency spectrum into two different functional portions (while also eliminating the need for a diplexer to do the spectral separating) and clears the way for programmable use of different portions of the active frequency spectrum, such as by means of a software-defined-radio (SDR) communications systems approach.

    Reply
  43. Tomi Engdahl says:

    The Wireless World Converges at Mobile World Congress 2018
    Promising products and critical concerns meet in Barcelona.
    http://www.mwrf.com/community/wireless-world-converges-mobile-world-congress-2018?NL=MWRF-001&Issue=MWRF-001_20180403_MWRF-001_159&sfvc4enews=42&cl=article_1_b&utm_rid=CPG05000002750211&utm_campaign=16378&utm_medium=email&elq2=e326e0f2a820412eaf74ac722e8de849

    Barcelona, Spain became the focus of the mobile world once again as players from every part of the mobile/wireless ecosystem converged for the world’s largest mobile communications conference. The annual Mobile World Congress (MWC) provides companies large and small the opportunity to compete for the attention of potential customers, prospective partners, and media from around the world.

    Reply
  44. Tomi Engdahl says:

    1G to 10G Ethernet Dynamic Switching Using High-Speed Serial I/O Solution in the UltraScale Architecture
    https://www.eeweb.com/app-notes/1g-to-10g-ethernet-dynamic-switching-using-high-speed-serial-i-o-solution-in-the-ultrascale-architecture

    This application note targets the Ethernet designs that require dynamic switching between 1Gb/s to 10Gb/s using high-speed serial I/O links. The design uses the Xilinx Ethernet Solution Suite along with a Xilinx GTH/GTY Transceiver to form the Ethernet interface. The same GT is used to interface with the gigabit Ethernet Physical Coding Sublayer/Physical Media Dependent (PCS/PMA) and the 10G Ethernet PCS/PMA IP core. The rate switching is handled using the Dynamic Reconfiguration Port (DRP) of GTs.

    The figure displays the reuse of the same transceiver for both 1G and 10G protocols. The GT channel (hard block) is separated from the PCS/PMA IP core and is configured using the DRP controller.

    The Ethernet packet is transferred via 1G MAC – 1G PCS/PMA and arrives at input 1 of the 2:1 MUX, while the 10G packet arrives at input 0 of the MUX. The user-driven select line of the MUX controls the routing of the Ethernet packet to the GT interface.

    https://www.eeweb.com/app-notes/download/180925

    Reply
  45. Tomi Engdahl says:

    ECI releases NPT-1300 packet-optical transport platform for improved end-to-end metro service delivery
    http://www.lightwaveonline.com/articles/2018/04/eci-releases-npt-1300-packet-optical-transport-platform-for-improved-end-to-end-metro-service-delivery.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-04-05&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2058550

    Optical transport systems vendor ECI is using the MPLS+SDN+NFV World Congress in Paris, April 10-14, to launch its new NPT-1300 multiservice packet-optical transport platform to simplify end-to-end metro service delivery. Incorporating carrier-grade assurance and visibility with packet efficiency and multiservice support, the NPT-1300 delivers the scalability and the service agility required to meet increasing traffic demands in the metro, says ECI.

    ECI asserts the NPT-1300 is ideal for mobile backhaul (3G, 4G, and 5G), wholesale service delivery, residential multiplay, metro aggregation, and business VPN connectivity services. The platform offers multiservice, carrier-grade redundancy, service assurance, and Elastic MPLS (which ECI says is a combination of both IP and TP) and segment routing. ECI’s NPT-1300 also features real-time control through the ECI Muse software suite, and several service and networking applications for streamlined service delivery and automate network operations.

    Reply
  46. Tomi Engdahl says:

    Ekinops PM 200FRS02 FlexRate line module deployed in Epsilon greenfield network
    http://www.lightwaveonline.com/articles/2018/04/ekinops-pm-200frs02-flexrate-line-module-deployed-in-epsilon-greenfield-network.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-04-05&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2058550

    Ekinops (Euronext Paris – FR0011466069 – EKI) says that Epsilon is deploying its new PM 200FRS02 flexible rate line module in a greenfield network built by the privately owned global communications service provider.

    According to Ekinops, need for Epsilon’s software-defined networking (SDN) and on-demand connectivity has increased over the past 18 months. Epsilon is addressing this growing demand by building a greenfield network in New York City. The fiber network will use Ekinop’s PM 200FRS02 line modules, which offer as much as 200G of bandwidth for connectivity of 100GbE services to any of Epsilon’s 18 points of presence (PoPs) in the New York metro area.

    To create 100G or 200G transport links, the PM 200FRS02 uses dual QSFP28 client ports with software-selectable line side modulation that can be provisioned as either DP-QPSK or 16QAM, says Ekinops.

    Epsilon chose Ekinops for the advanced performance and flexibility offered by the PM 200FRS02, as well as its scalability, says the optical transport systems specialist. As many as six modules can be deployed in a single shelf providing 1.2 Tbps of capacity in just 2RU.

    “In a global market driven by SDN, it is critical that we deliver new levels of scalability and flexibility. Our partners want on-demand solutions that can grow with their businesses no matter where they are located,”

    Reply
  47. Tomi Engdahl says:

    COBO offers Release 1.0 specifications for onboard optical modules
    http://www.lightwaveonline.com/articles/2018/04/cobo-offers-release-1-0-specifications-for-onboard-optical-modules.html?cmpid=enl_lightwave_lightwave_enabling_technologies_2018-04-05&pwhid=6b9badc08db25d04d04ee00b499089ffc280910702f8ef99951bdbdad3175f54dcae8b7ad9fa2c1f5697ffa19d05535df56b8dc1e6f75b7b6f6f8c7461ce0b24&eid=289644432&bid=2058550

    Three years after its founding, the Consortium for On-Board Optics (COBO) used OFC 2018 in San Diego last month as a stage on which to debut its long-awaited Release 1.0 specifications. The specifications cover three classes of onboard optical modules, backed by a pair of electrical interfaces. The optical modules will support transmission rates up to 800 Gbps. COBO expects the modules to meet a range of data center network requirements, including those that would require coherent transmission.

    The specification defines requirements for electrical, mechanical, optical, and thermal parameters for embedded optical modules. The specification supports optical interfaces based on the recently completed IEEE 802.3bs-2017 standard, which covers both 400 Gigabit Ethernet and single-mode 200 Gigabit Ethernet applications.

    Release 1.0 defines each of the three module classes in combination with the two electrical interfaces – one eight lane, the other sixteen lane, based on 25-GBaud PAM4 per lane – to meet both 1x400G and 2x400G requirements. The latter can be in either independent or integrated format. The three classes feature a common host connector and width of 20 mm for the x8 configuration and 36 mm for the x16 form factors. The three classes vary in length, with Class A at 30 mm, Class B at 40 mm, and Class C at 60 mm. The varying lengths provide the necessary packaging room to contain the optics and electronics necessary to meet targeted reach requirements.

    The resulting modules could be applied to a variety of applications. COBO has developed a reference application that details a 1RU 12.8-Tbps data center switch with either front-to-back or back-to-front airflow that was used to develop the specifications.

    Reply
  48. Tomi Engdahl says:

    COBO offers Release 1.0 specifications for onboard optical modules
    The Release 1.0 specification can be downloaded from http://onboardoptics.org/.

    Reply
  49. Tomi Engdahl says:

    How To Build A Modern Network
    https://semiengineering.com/how-to-build-a-modern-network/

    How to sort through the different telemetry approaches for the optimum solution.

    There have, in recent years, been fundamental changes to the way in which networks are implemented, as data demands have necessitated a wider breadth of functionality and elevated degrees of operational performance. Accompanying all this is a greater need for accurate measurement of such performance benchmarks in real time, plus in-depth analysis in order to identify and subsequently resolve any underlying issues before they escalate.

    A broad spectrum of different options are available when looking to extract telemetry material, whether that be passive monitoring, active measurement, or a hybrid approach. An increasingly common practice is the piggy-backing of telemetry information onto the data packets that are passing through the network. This tactic is being utilized within both in-situ OAM (IOAM) and in-band network telemetry (INT), as well as in an alternate marking performance measurement (AM-PM) context.

    WHITE PAPER
    Network Telemetry Solutions for Data Center and Enterprise Networks
    https://www.marvell.com/documents/8ni0aekscuc444ay31nn/

    Active measurement uses dedicated control plane (OAM) messages. The performance of these messages is monitored, thereby giving an indicator ofthe performance of the user traffic.

    Reply

Leave a Reply to Tomi Engdahl Cancel reply

Your email address will not be published. Required fields are marked *

*

*