Tuesday, March 27, 2018

AT&T outlines plans for Software-Defined Network in 2018 and beyond

AT&T, which has previously stated plans to virtualize and software-control 75% of its core network functions by 2020, said that it is on target to reach the 65% virtualization milestone this ahead, coinciding with the rollout of mobile 5G service.

In a blog post, Chris Rice, Senior Vice President – AT&T Labs, Domain 2.0 Architecture and Design, outlines several major open source initiatives that the company has been spearheading to drive network transformation.

  • ONAP – The operating system for Network Clouds. Since formation a year ago, the project now brings together over 50 of the largest network and cloud operators and technology providers from around the globe, representing more than 60% of the world’s mobile subscribers. AT&T is leading the software contribution and together with project members plans to deliver the second ONAP release, Beijing, in the second quarter of this year.
  • DANOS – The operating system for individual white box servers that power a virtualized network. And we recently announced we are taking white boxes, running on DANOS, to scale in our network and will plan to install as many of 60,000 of these white box routers to support our 5G build out over the next few years.
  • Acumos – An industry-first AI platform and marketplace, co-created with Tech Mahindra, that makes it easy to chain multiple microservices together in a simple drag-and-drop interface. And The Linux Foundation recently announced the public availability of Acumos, meaning anyone can now access the platform and begin building AI applications. We’re also releasing our Acumos whitepaper, where we highlight our direction and efforts in this area.
  • Akraino Edge Stack – A complete software platform for edge computing systems and applications. The Linux Foundation also recently announced that Intel among others have signed on as members of the community. We’re seeing early indicators of progress that are encouraging, and this growing collaboration will help to expedite the maturity and adoption of edge cloud.


AT&T also plans to roll out over 60,000 white box routers over the next several years across the U.S. to enable advanced services on its mobile 5G infrastructure. These white box routers run the “Disaggregated Network Operating System,” or dNOS, which uses technology from its acquisition of Vyatta.

"White box represents a radical realignment of the traditional service provider model,” said Andre Fuetsch, chief technology officer and president, AT&T Labs. “We’re no longer constrained by the capabilities of proprietary silicon and feature roadmaps of traditional vendors. We’re writing open hardware specifications for these machines, and developing the open source software that powers these boxes. This means faster hardware upgrades, since anyone can build to these specs. And software upgrades that move at internet speed. We’re doing this all while keeping costs low so we can focus on expanding our nationwide mobile 5G footprint for our customers as quickly as possible."

http://about.att.com/innovationblog/att_framework

FirstNet's Dedicated Evolved Packet Core Goes Live

AT&T confirmed that the dedicated evolved packet core for FirstNet is now live nationwide.

FirstNet is the country’s first nationwide public safety communications platform dedicated built by AT&T in public-private partnership with the First Responder Network Authority.

The FirstNet network core is built on physically separate hardware from the AT&T network. It provides first responders with their own separate, nationwide broadband network and it is being certified by the First Responder Network Authority. The packet core forms the basis for the unified, interoperable and nationwide communications system.

First responders on the FirstNet evolved packet core can now access "First Priority", which turns FirstNet’s always-on access to priority and preemption up a notch, giving first responders 2 more priority levels. With 3 options at their fingertips, public safety agencies will have the ability to assign priority levels based on their command structure or shifting needs. The launch also unlocks an Incident Management Portal, which allows official to uplift critical users to the highest priority levels. Using the Incident Management Portal, they can make these adjustments in near real time to help first responders.  Public safety can also temporarily uplift other non-first responder users that are essential to managing an incident – like utilities or transportation. Future public safety capabilities will include mission-critical Push-to-Talk, z-Axis location-based services, etc.

A controlled introduction is now underway. A number of FirstNet-ready devices have been announced, including popular mobile devices like the Samsung Galaxy S9/S9+, and rugged mobile devices like the Sonim XP8 and XP5s (coming soon). The FirstNet SIM will also work with other devices, but firmware updates may be required.

The NETGEAR Nighthawk M1 Mobile router – a mobile hotspot router – is another product that will accept the FirstNet SIM.

“This is what public safety has spent years advocating for,” said Scott Edson, executive director, Los Angeles Regional Interoperable Communication System (LA-RICS). “We knew giving first responders a network that they could truly call their own was possible from our work on LA-RICS – 1 of 5 FirstNet early builder projects. But to see public safety’s network core roar to life nationwide, well, there are no words for how meaningful that is. We at LA-RICS look forward to connecting our sites to the FirstNet network core.”

“Outdated communications capabilities are a threat to public safety. We’ve seen it repeatedly when disasters strike – from September 11, the Boston Marathon and Parkland,” said Ed Davis, former Boston police commissioner. “We’ve been advocating for the future of communications to bring us a modern solution that will empower us with next generation tools. And with FirstNet, that future is here, giving us an experience we can’t get on any commercial network.

OpenContrail is rebranded as "Tungsten Fabric"

OpenContrail, which is the SDN framework originally created by Juniper Networks and open sourced in 2013 and which is now hosted by The Linux Foundation as an open-source network virtualization platform for the cloud, has changed its name to Tungsten Fabric.

Tungsten Fabric provides a single point of control, observability and analytics for networking and security. It is integrated with many cloud technology stacks, including Kubernetes, Mesos, VMware and OpenStack. It supports private cloud, hybrid cloud and public cloud deployments such as AWS and GCE. Tungsten Fabric includes a high-performance vRouter that connects container, VM and bare-metal applications, and a controller which orchestrates network overlays, switch fabrics and router gateways.

The Linux Foundation said Tungsten Fabric is now deployed at massive global scale, across public and private IaaS, CaaS and PaaS environments ranging from hyperscale cloud providers and telecom operators to enterprises.

Tungsten Fabric’s mission is "to build a ubiquitous, easy-to-use, scalable, secure and cloud-grade SDN stack that provides a network fabric capable of connecting diverse environments."

Contributors and community members include Aricent, AT&T, Bell, Cavium, CertusNet, CloudOps, CodiLime, Intel, Juniper Networks, Lenovo, Mellanox, Mirantis, Netronome, Orange, SDN Essentials, TechTrueUp, and Yandex.

“We are pleased to create Tungsten Fabric with a neutral governance under The Linux Foundation,” said Arpit Joshipura, general manager, networking, The Linux Foundation. “The set up allows Tungsten Fabric to collaborate with other Linux Foundation and Networking projects. We’re looking forward to expanded collaboration across a growing software-defined ecosystem.”

https://tungsten.io

The DANOS Project promises a Unified Network Operating System

The Linux Foundation will host a new Disaggregated Network Operating System (DANOS) project to enable community collaboration across network hardware, forwarding and operating system layers.

DANOS is initially based on AT&T’s “dNOS” software framework of an open, cost-effective and flexible alternative to traditional networking operating systems.

A first code release is expected the second half of 2018.

“We are pleased to welcome DANOS to The Linux Foundation community of open, collaborative innovation,” said Arpit Joshipura, General Manager of Networking, The Linux Foundation. “DANOS will provide an open NOS framework that leverages existing open source resources and complementary platforms such as switches and white box routers. We invite others in the broader ecosystem to join the effort to accelerate innovation and creation of an industry-standard disaggregated NOS.”

“As far as we know, DANOS is an industry first: an open-source, carrier-grade operating system for wide area networks,” said John Medamana, Vice President, Packet Optical Network, AT&T. “DANOS is a milestone for us and the industry, and we’re excited to see how developers and other users implement and build upon it.”

“ONF has been a champion of disaggregation, white boxes and open source and in this regard, we are happy to see DANOS launched as a carrier-grade, open source network OS for white boxes,” said Guru Parulkar, Executive Director, ONF. “We are looking forward to DANOS and ONF’s recent open source platform Stratum, thin switch OS, working together.”

AT&T’s whitepaper on the dNOS framework is available here: http://about.att.com/content/dam/innovationblogdocs/att-routing-nos-open-architecture_FINAL%20whitepaper.pdf

NVIDIA brings 10x performance for Deep Learning System

At its annual GTC conference in San Jose, NVIDIA introduced its 2 petaflop, DGX-2 Deep Learning System, promising a 10x performance boost on deep learning workloads compared with the previous generation from six months ago.

Key advancements in NVIDIA platform include a 2x memory boost to NVIDIA Tesla V100 datacenter GPU, and a revolutionary new GPU interconnect fabric called NVIDIA NVSwitch, which enables up to 16 Tesla V100 GPUs to simultaneously communicate at a record speed of 2.4 terabytes per second. NVIDIA also introduced an updated, fully optimized software stack.

These advancements enable the NVIDIA DGX-2 server to deliver two petaflops of computational power -- the equivalent of 300 servers occupying 15 racks of datacenter space, while being 60x smaller and 18x more power efficient.

“The extraordinary advances of deep learning only hint at what is still to come,” said Jensen Huang, NVIDIA founder and CEO, as he unveiled the news at GTC 2018. “Many of these advances stand on NVIDIA’s deep learning platform, which has quickly become the world’s standard. We are dramatically enhancing our platform’s performance at a pace far exceeding Moore’s law, enabling breakthroughs that will help revolutionize healthcare, transportation, science exploration and countless other areas.”

Verizon to offer public safety private core

Verizon announced plans to offer its own public safety private core network.

The dedicated private core will enhance Verizon’s 4G LTE network and be tuned for public safety uses. Additionally, Verizon offers public safety customers preemption and mobile broadband priority service at no additional charge.

Verizon said its public safety private core provides several key features to public safety customers including traffic segmentation, priority and preemption, improved security, and enhanced service management and control. It is connected to Verizon’s Radio Access Network (RAN) which utilizes spectrum in various bands including the 700 MHz, 800 MHz Cellular, 1.9 GHz PCS, and 1.7/2.1 GHz AWS bands. The public safety core separates data traffic of public safety mobile users from commercial users across Verizon’s 4G LTE network. Public safety users will have their data immediately recognized as public safety with priority access at the tower and through the network. The private core leverages leading edge networking technology to provide security, flexibility and reliability.

“Public safety answers the call when we need them most,” said John Stratton, Verizon executive vice president and president of global operations. “We remain committed to providing them innovative communications solutions that help them help us, and we are honored by the trust they put in Verizon every day.”

Verizon's public safety private core will be generally available on March 29, 2018.

Ciena adds ONAP elements to its Blue Planet Platform

Ciena announced enhanced policy capabilities into its Blue Planet intelligent automation platform that incorporates the architectural framework from the Linux Foundation’s open-source Open Networking Automation Platform (ONAP) to aid network providers’ evolution to more adaptive, software-centric networks. Specifically, 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.

Ciena said its new Blue Planet policy subsystem will support advanced closed-loop automation use cases, such as preventing a potential failure or dynamically scaling capacity to support growing bandwidth demands.

“The future of networking lies with openness and the concept of ‘choice’, allowing network operators to select best-of-breed technologies and solutions. These Blue Planet enhancements bring us closer towards the common goal of software-led, policy defined, network transformation. They also help forge a path for more adaptive networks that can self-learn, self-configure, self-heal, and self-optimize by constantly assessing network status,” stated Steve Alexander, Chief Technology Officer, Ciena. 

Digital Realty maintains Five Nines data center reliability

Digital Realty has achieved “five nines” of uptime for its data center suites for the 11th consecutive year, exceeding 99.999 percent availability throughout 2017.

Digital Realty recently surpassed 1.7 billion operating minutes across its 205 datacenters, encompassing approximately 27 million square feet in 33 metropolitan areas globally. Over the past 11 years, the number of data center suites operated by Digital Realty has increased from 74 suites at the start of 2007 to over 750 by the end of 2017.

“Achieving ‘five nines’ of uptime for the 11th year demonstrates our commitment to resiliency and to being a true business partner to all of our customers, from enterprises to telecommunications customers and cloud service providers,” said A. William Stein, Digital Realty’s Chief Executive Officer. “We are especially proud of this milestone given our rapid growth and remain focused on ensuring that all of our facilities meet the same stringent standards for design, construction and operations.”

Marvell integrates its Automotive Ethernet Switch into NVIDIA DRIVE

Marvell confirmed that its 88Q5050 secure automotive Ethernet switch is integrated into the NVIDIA DRIVE Pegasus platform for autonomous vehicles.

Marvell said its secure switch can handle multi-gigabit applications for OEM car manufacturers to deliver an in-car network that supports sensor fusion, cameras, safety and diagnostics. Its switch design employs a deep packet inspection (DPI) engine and trusted boot functionality to ensure a robust level of security. The switch also supports both blacklisting and whitelisting addresses on all its Ethernet ports to further enhance its security especially against denial of service attacks.

The NVIDIA DRIVE Pegasus computing platform has been designed to handle Level 5 driverless vehicles and compute over 320 trillion operations per second. The platform combines deep learning, sensor fusion, and surround vision to understand what’s happening around the vehicle in real-time. The Marvell 88Q5050 layer 2 managed secure IP switch enables the fast and safe transmission of this data for next generation connected vehicles.

“The 88Q5050 is Marvell’s latest addition to the automotive portfolio of wired and wireless network solutions and is designed to prevent malicious attacks or compromises to data streamed in and out of the vehicle. Marvell has over 20 years of Ethernet IP experience and this partnership with NVIDIA demonstrates our continued commitment to innovating and leading in automotive technology. Marvell is excited about the future of automotive and will utilize its extensive portfolio to accelerate the adoption of automotive Ethernet,” said Thomas Lagatta, executive vice president of Sales and Marketing at Marvell.

Oracle debuts Autonomous Database in the Cloud - 1/2 cost of AWS

Oracle Executive Chairman and CTO Larry Ellison introduced the first service based on new Oracle Autonomous Database -- a self-managing, self-securing, self-repairing database cloud service called Oracle Autonomous Data Warehouse Cloud.  The service uses machine learning to deliver performance, security capabilities, and availability with no human intervention, at "half the cost of Amazon Web Services."

The warehouse provisioning service spins up a secure data warehouse with automatic backup, encryption, and a high availability architecture in seconds. The company says its Oracle Autonomous Data Warehouse Cloud is so fast that guarantees the same workload at half the cost of Amazon Web Services.

Oracle Autonomous Data Warehouse Cloud delivers all of the analytical capabilities, security features, and high availability of the Oracle Database without any of the complexities of configuration, tuning, and administration -- even as warehousing workloads and data volumes change.

"This technology changes everything," said Ellison. "The Oracle Autonomous Database is based on technology as revolutionary as the Internet. It patches, tunes, and updates itself. Amazon's databases cost more and do less."


See also