Tuesday, June 13, 2017

Cisco: expansion of the IP universe is accelerating

Cisco: expansion of the IP universe is accelerating

Cisco has just published its latest Visual Networking Index, an annual report that uses network traffic data from its service provider customers to forecast usage patterns for the next four years. Cisco has been publishing these VNI reports for 12 years and over that time has attracted quite a following. The reports are often cited in the media and used by many other network operators, regulators and Internet companies for planning purposes.

The first question on everyone's mind is 'how fast is the Internet growing?' Cisco's answer for this year, along with its responses for the past few years, are as follows:

·         2017: Global IP traffic is expected to increase three-fold from 2016 – 2021, reaching an annual run rate of 3.3 zettabytes by 2021, up from 1.2 zettabytes in 2016. Busy hour Internet traffic is increasing faster than average Internet traffic, and busy hour Internet traffic will grow 4.6-fold (35% CAGR) from 2016 to 2021, reaching 4.3 Pb/s by 2021, compared to average Internet traffic that will grow 3.2-fold (26% CAGR) over the same period to reach 717 Tb/s by 2021.

·         2016: Global IP traffic to nearly triple at a CAGR of 22% over the next five years (2015-2020) as more than a billion new Internet users come online and new applications take hold.

·         2015: Global IP traffic to triple between 2014 and 2019, when it will reach a record 2 zettabytes. This equates to a CAGR of 23% and marks the first global CAGR increase in consecutive VNI forecasts in nearly a decade.

·         2014: Global IP traffic to grow three-fold from 2013 to 2018, reaching 1.6 zettabytes annually by 2018, representing a 21% CAGR over the forecast period.

·         2013: Global IP traffic to reach 1.4 zettabytes (23% CAGR from 2012 to 17).

·         2012: Global IP traffic forecast to be 1.3 zettabytes by 2016.

In short, this year's Cisco's VNI predicts a 26% CAGR for global IP traffic over the next five years, which means that growth is not just ripping along but actually accelerating. The universe is expanding quickly and accelerating. It is an important observation with implications for everyone in the network ecosystem.

Looking at these big numbers for overall Internet traffic, and knowing that video represents such a significant percentage of that 26% CAGR, one cannot help but wonder why Content Delivery Networks (CDNs) would not have mitigated more of this traffic load. Considering the load an all-day Netflix binge on a 4K monitor, if tens of millions of viewers in the same cities and neighbourhoods are engaged in similar behaviour and watching the same shows, how effective are CDNs? The Cisco VNI numbers indicate that CDN traffic is growing even faster! A 44% CAGR globally and even faster in many regions.

Cable provider IP traffic is a different category

One might also think that the 26% CAGR for public IP traffic would be partially due to the migration or planned migration of cable operators to IP delivery platforms. But that is not the case. Cisco VNI has a separate category for Managed IP video, which is defined as IP traffic generated by traditional commercial TV services. It is not considered Internet traffic because its remains within the footprint of a single service provider. Below are the numbers for that category.






Infinera Intros XTM II for Cloud Scale Metro Packet-Optical

Infinera introduced its next generation packet-optical platform for metro networks.

The XTM II, which delivers Layer 0, Layer 1 and Layer 2 services, is optimized for bandwidth-intensive cloud scale applications at the metro edge, such as Remote PHY, 5G transport and data center interconnect (DCI).

Infinera said its new platform builds on its widely deployed XTM Series, but now adds 200 gigabits per second (200G) per wavelength capabilities, with an eightfold density increase and a reduction in power per gigabit of 3.5 times. Power consumption is believed to be the lowest in the industry for 100/200G transport. It also features Infinera Instant Bandwidth, which is the company's open grid line system with SDN control. This gives network operators a highly flexible, open and software-programmable packet-optical solution for Layer 0, Layer 1 and Layer 2 services.

A key component of the XTM II platform is the new range of 200G per wavelength traffic units, featuring:
  • The 400G Flexponder: A dual, 200G muxponder that uses 16QAM (quadrature amplitude modulation) for high-capacity transport, or a dual 100G transponder that uses quadrature phase-shift keying (QPSK) for longer reach operation. This device provides 400G of line and client capacity per slot, giving an eightfold density increase over the previous generation. Including optics, the device operates at as low as 20 watts per 100G service, which the company believes is the lowest power consumption per 100G available in the industry on any wavelength-division multiplexing (WDM)-based platform.
  • The 200G Muxponder: A 200G Layer 1 muxponder that supports a broad range of client signals, including 10G/40G/100G Ethernet and Optical Transport Network (OTN) as well as 8/16/32G Fibre Channel. The device can also be paired to create an OTN add-drop multiplexer (ADM).
  • The EMXP440 Packet-Optical Transport Switch: A high-capacity addition to the existing range of EMXP devices that provides Layer 2 packet-optical switching with dual 100/200G ports and 12 or 24 10G ports. The EMXP440 supports Carrier Ethernet (CE) and MPLS-TP, packet transport with sub-50 milliseconds protection, Metro Ethernet Forum (MEF) CE 2.0 service creation and quality of service-aware traffic aggregation. In addition, the EMXP440 has feature-harmonization with the EMXP/IIe range and PT-Fabric.
Additional enhancements to the XTM Series include:


  • A new portfolio of XTM II upgraded chassis for improved power management and cooling and increased density to support nodes that require large volumes of new traffic units. 

  • Instant Bandwidth capability, enabling the on-demand licensing of 100G bandwidth increments to align capital expense spend with service revenue and to reduce operational expenses through automated software activation of new capacity.
  • New 400G+ per wavelength-ready flexible grid 4x and 9x ROADM modules and optimized hybrid erbium-doped fiber amplifier (EDFA)/Raman optical amplifiers to support sophisticated modulation formats and higher baud rates required above 100G. In addition, the new XTM II open flexible grid line system supports fiber capacity up to 24 terabits per second.
  • A unified solution providing end-to-end software control from core to access. The XTM Series, including the XTM II, is supported by Infinera’s Xceed Software Suite and DNA network management system. This new range of packet-optical platforms provides network operators with leading low power and high density at Layer 0, Layer 1 and Layer 2, and supports full interworking with the large installed base of XTM Series and the DTN-X platforms.
“While backward compatibility is key as it allows current customers to take advantage of over 30,000 existing XTM Series chassis, what is more significant are the capabilities XTM II brings to new networks deployed by both new and existing customers,” said Karl Thedéen, Infinera Senior Vice President, Head of Metro Business Group.

The XTM II chassis are already shipping to customers and the 400G Flexponder will be available in Q3, with additional components of XTM II available by the end of 2017.

https://www.infinera.com/xtm-ii-cloud-scale-metro-packet-optical-applications/

BT Labs demonstrates multi-400 Gbit/s superchannel

BT has announced a new 400 Gbit/s single-carrier based technology solution for the transmission of large volumes of data over core optical networks developed by researchers at the BT Labs in Adastral Park, Ipswich in the UK.

BT noted that the trial of the technology builds on its previously demonstrated Superchannel concept, which achieved transmission speeds of 5.6 Tbit/s in 2016 by combining 200 Gbit/s wavelengths of light into a single optical fibre.

For the latest demonstration BT, working with partner Huawei, has developed a technique that combines multiple 400 Gbit/s wavelengths over a single fibre. Using the new technique, BT researchers believe that transmission speeds of more than 13 Tbit/s can be achieved using the same amount of light spectrum as in the demonstration last year.

BT's latest Superchannel concept is based on the ultra-efficient use of light carried over optical fibre, termed spectral efficiency. The trial demonstrated that it is possible to transmit at multi-terabit speeds over existing core networks with stable, long term, error free performance and 6.25 bits/s/Hz spectral efficiency. BT claims that this spectral efficiency is the highest achieved on a real-world fibre link using production-grade hardware and software.

For the demonstration the technology was trialled on a live optical loop between BT Labs at Adastral Park and BT's Bishops Stortford exchange over a distance of 250 km. BT is demonstrating the technology at its Innovation Week 2017 event at Adastral Park.



  • In May last year, BT announced it had successfully transmitted speeds of 5.6 Tbit/s over a single optical fibre, and also that working with Huawei had demonstrated speeds of 2 Tbit/s over a live core network spanning 700 km between London and Dublin. The demonstration of 5.6 Tbit/s on an optical superchannel over a closed loop comprised 28 x 200 Gbit/s (64 GBaud/QPSK) sub-channels, bundled together to provide combined capacity with high spectral efficiency.

China Mobile to deploy Nokia home gateways for 30m users

Nokia and China Mobile announced the deployment of millions of home gateways to provide residential customers in 29 provinces in China with access to fibre-based ultra-broadband applications and intelligent home services.

Under the agreement, China Mobile will deploy home gateway units featuring Nokia's solution to over 30 million users during 2017. Utilising established FTTH networks that enable gigabit speeds for end customers, deployment of the new gateway is intended to extend Internet coverage within the home and enable IoT communications between devices and sensors.

Additionally, the ability to flexibly add software functions and enhanced analytics capabilities will allow China Mobile to deploy and deliver a new intelligent home experience and associated services.

Nokia noted that the latest contract with China Mobile extends a long-term fixed networks partnership between the companies that also encompasses the development of the telco's GPON network to support mobile backhaul.
According to market research company IDATE, China Mobile is expanding its position as a converged telecom operator and currently serves more than 31 million FTTH subscribers. It noted that the operator is leveraging its extensive fibre access network to deliver ultra-broadband applications such as 4K TV and gigabit access to customers in a number of provinces. The addition of intelligent home gateway technology is expected to enable China Mobile to differentiate its services.



  • Earlier this year, Nokia announced that its Nuage Networks venture had been awarded a contract to supply its Virtualized Services Platform (VSP) for China Mobile's first commercial public cloud project. For the project China Mobile is using Nuage's scalable SDN solution for a deployment of approximately 2,000 public cloud servers in Beijing and Guangzhou.
  • Nuage Networks' VSP solution is designed to enable China Mobile to virtualise its multi-tenant data centre networks and establish connectivity among computing resources while also delivering new features to customers. China Mobile had previously deployed Nuage Networks SDN technology in CMCC's DevOps private cloud architecture.

TE SubCom selected for South Pacific cable maintenance

TE SubCom, a TE Connectivity company and leading provider of submarine communications technology, announced that it has been awarded the South Pacific Marine Maintenance Agreement (SPMMA), a five-year service agreement with 14 cable operators in the region.

Under terms of the agreement, which began in March 2017, SubCom will maintain more than 51,000 km of cable that comprise 19 disparate telecommunications and power cable systems.

The SPMMA area covers the South Pacific region from Singapore in the west to Tahiti in the east, and from the southernmost point of New Zealand to Hawaii in the north. The agreement involves the maintenance cable systems across more than 28 million sq miles of the Pacific Ocean and will be supported by the cable ship Reliance, which is based in the South Pacific region.

As part of the agreement, SubCom will leverage its modern marine assets and technology portfolio to enable the delivery of maintenance services throughout the region.


  • TE SubCom is currently engaged in deploying the Hawaiki Submarine Cable, a 14,000 km trans-Pacific cable system that will link Australia and New Zealand to the mainland U.S. and Hawaii and American Samoa. New Zealand's Hawaiki Cable announced in 2013 that it had awarded a turnkey supply and installation contract to TE SubCom to build the cable system.
  • The 100 Gbit/s-enabled Hawaiki cable system will support capacity of up to 10 Tbit/s per fibre pair on the Australia/New-Zealand to U.S. route, while a number of Pacific Islands along the route will be able to connect to the main trunk.

Netronome enhances Agilio SmartNIC for OPNFV Danube

Netronome, a provider of intelligent networking solutions, has announced support for the latest OPNFV software release, OPNFV Danube, and integration into its Agilio SmartNIC platform, which is claimed to improve networking performance by up to 4x and to reduce CPU consumption by over 5x.

Netronome noted that at the OPNFV Summit in Beijing, Open-NFP.org is hosting the OPNFV Data Plane and VNF Acceleration Mini-Summit, which will feature technical sessions focused on acceleration of VNFs within the framework of OPNFV Danube-supported data planes such as Open vSwitch (OVS) and the Fast Data project (FD.io). Participants in the sessions include China Mobile Research Institute (CMRI), Guangzhou Research Institute at China Telecom, OPNFV, Open-NFP, SK Telecom, Spirent Communications and ZTE.

Netronome explained that designers and operators of cloud and telco data centre networks recognise the need for server acceleration and offload to realise the TCO benefits of NFV. SmartNICs can play a key role in efficiently delivering acceleration, while acceleration and offload of portions of VNF functions can offer further cost benefits.

During the OPNFV Data Plane and VNF Acceleration Mini-Summit, OPNFV community members are proposing an open framework and API to cover NFVi and VNF acceleration designed to promote broad cooperation and collaboration in the industry. The minis-summit is hosted by Open-NFP.org, a global, community-driven organisation that seeks to enable open, collaborative development in the area of data plane acceleration for network functions processing in server networking.

Open-NFP hosts 40-plus open projects and comprises over 100 developers, including service providers, OEMs, research institutions and academia. The portal also offers developers support via a P4- and C-based data plane software development environment and open source code related to Netronome Agilio SmartNICs.

Sessions to held during the event will cover areas including:

1.         NFVi acceleration models and offload architectures for OVS and VPP data planes using SmartNICs.

2.         VNF acceleration models and offload architectures for NFV using SmartNICs.

3.         An open API model for enabling NFVi and VNF acceleration, including the use of sandbox functions based on the P4 and/or C programming languages.

4.         A proposal for developing and testing VNF acceleration under the umbrella of the OPNFV Pharos Labs initiative, leveraging resources and support from the Open-NFP community.

Regarding the demonstration, Nick Tausanovitch, VP of solution architecture at Netronome, said:

-           "Enhancements to OVS-DPDK and introduction of new data plane technologies such as FD.io improve networking performance, however, these technologies rely on x86 compute cycles for networking functions, bringing down server efficiency while starving VNFs of bandwidth and reducing the CPU cycles available".


-           "The Agilio SmartNIC platform from Netronome, integrated with the OPNFV Danube software platform release, relieves such bottlenecks, enabling service providers to deliver new 4G and 5G services faster and more economically".



  • At Mobile World Congress in February, Netronome announced that its flagship Agilio CX SmartNIC platform has been integrated into a range of NFVi solutions from OEMs, including Juniper Networks, Mirantis and Nuage Networks (Nokia) to help telcos prepare their data centre infrastructure for the growth in data and new services.
  • Netronome also announced that Ericsson had added support for the Agilio CX SmartNIC platform within its NFVi solution to enable offload and accelerated datapath processing and to keep virtual machines (VMs) and VNFs hardware independent for rapid onboarding.

OFS expands InvisiLight fiber portfolio

OFS, a manufacturer of advanced fibre optic products and a Furukawa Electric company, has launched the InvisiLight Façade Solution for low-rise buildings or garden-style dwelling units, expanding the InvisiLight portfolio that also includes the drop cable solution to residential homes, MDU solution for building hallways and ILU solution for indoor living units.

OFS' InvisiLight portfolio also integrates with the SlimBox branded portfolio of terminals and closures to provide a modular solution offering for a range of market demands.

The new InvisiLight Façade solution is offered with 12 or 24 fibre-count indoor/outdoor rated cable, which can be installed on the face of a building or placed behind outdoor structures such as rain water downspouts. The 12 and 24 fibre cables have diameters of 3.0 mm and 3.8 mm, respectively, and can be either factory pre-terminated or connectorised on site with a fusion splice or mechanical connector.

During the installation process, the cables are connected to an outdoor building terminal typically installed at the base of a building. From the terminal, the cable can be installed vertically on the face of the building into the attic and then connected to a distribution terminal from which EZ-Bend optical cables can be run into living units.

Alternatively, the cable can be placed along the building facade using clips to pass each living unit, with compact slack loops placed outside or inside the unit. From there the optical fibre can be extracted to connect to the other InvisiLight solutions to connect to the optical network terminal (ONT).

OFS stated that its established EZ-Bend optical fibre is used in the InvisiLight Façade cable products, as well as the other InvisiLight portfolio of products. The EZ-Bend fibre features a 2.5 mm bend radius, which allows it to accommodate the sharp corners often encountered when fitting the cable onto buildings. OFS noted that the new InvisiLight Façade solution is designed to offer an almost invisible cable for installation onto the exterior of and within buildings.

Dell'Oro reports white box server shipments up 41% yr/yr in Q1

According to the latest Server Quarterly Report from Dell'Oro Group, white box server shipments continued to grow at a rapid pace in the first quarter, increasing 41% year on year, with the first quarter growth led principally by Google and Amazon, although Facebook and Microsoft are expected to increase their spending in the second quarter of 2017.

Highlights from Dell'Oro's server report for the first quarter of 2017 include:

1.         Overall cloud data centre expenditure is projected to remain strong for the remainder of the year, although it is forecast to be uneven during the year across the major cloud providers Google, Microsoft, Amazon and Facebook.

2.         The cloud providers are expected to continue to build-out data centres at a high rate and to invest in the latest server technologies based on the Intel Purley platform and 25 Gbit/s single-lane SerDes technology.

3.         Most of the major U.S.-based branded vendors, led by Hewlett-Packard Enterprise (HPE) and Dell Technologies, experienced quarter over quarter and year on year shipment declines due to a number of factors, including server migration from the enterprise/on premise to the cloud, typical first quarter weakness, and a pause in server purchases ahead of the Intel Purley server refresh cycle, which is expected in the second half of 2017.

Commenting on the report, Baron Fung, senior business analysis manager at Dell'Oro Group, noted, "Dell'Oro believes that most of the growth in the server market will come from cloud deployments… while the upcoming Intel Purley refresh cycle may spur overdue spending in enterprise, it will be the cloud providers that will set the market growth pace in 2017."

China Telecom Shanghai selects Huawei for gigabit network

Huawei announced that it will help the Shanghai Branch of China Telecom to deploy a gigabit network featuring 10 Gbit/s PON optical network terminals (ONTs) to support smart home services.

Shanghai Telecom is planning to build what is believed to be the first commercial FTTH network in China using 10 Gbit/s PON technology as it progresses towards providing full fibre coverage enabling 1 Gbit/s bandwidth in Shanghai over the next 3 years. This project is expected to make Shanghai the first gigabit city in China.

Huawei noted that Shanghai Telecom established a 3-year goal of moving from 100 Mbit/s to 1 Gbit/s in 2016, and is a leading company in the construction of 10 Gbit/s communities and delivery of 1 Gbit/s bandwidth to households.

By the end of 2016, Shanghai Telecom was providing 1 Gbit/s access for 269 communities. By the end of 2018, the average access rate of Shanghai Telecom's network is expected to rise from 50 to 280 Mbit/s, while user-perceived download rates are expected to rise from 13 to 100 Mbit/s.

As part of its gigabit services offering, Shanghai Telecom has released a range of home broadband services, including multi-channel 4K, video calling, video conferencing and other smart home services.

To address Shanghai Telecom's requirements, Huawei is supplying its large-capacity distributed optical line terminal (OLT) MA5800 and next-generation smart 10 Gbit/s PON ONT. The single sub-rack solution is able to support streaming of UHD 4K videos for 16,000 households concurrently, as well as offering support for 8K video.

The Huawei OLT is designed to enable gigabit convergence through multiple media, allow different services to share the same platform during cloud evolution and to support a large number of physical connections for smart home applications.


The ONT supports multiple Gigabit Ethernet ports and can bear simultaneous multi-channel 4Kvideo, video calling and virtual reality (VR) services. Additionally, Huawei is providing an open intelligent platform designed to flexibly support a range of smart home services.

The pieces are coming together at Dell Technologies – part 3

Wrapping up the coverage of last week's Dell EMC World, a key point to consider is how the many component companies now assembled will fit together and what strategic advantage this provides over the likely competition.

At the low end, Dell Technologies will face challenges from the 'white' competitors - vendors can supply generic PCs, servers, switches, etc. on thin margins. even single digit margins. Especially as the industry turns to open networking software and Open Compute Platform (OCP) hardware designs, Dell EMC could find itself in the same race to the bottom in margins. This is the race that drove IBM to sell off its PC and server business to Lenovo.

Looking around the exhibits and customer case studies at last week's event, it looks like there is a solid case that Fortune 500 enterprises are sticking with Dell. They simply have too much at stake to risk being out of date with their technology or going with a dodgy vendor. Just looking at the many victims for the WannaCry ransomware attacks, once again we can hear the experts reminding everyone to keep their systems up-to-date. Businesses must realise that running Windows XP on PCs that are six years old can put the whole enterprise at risk. As a top brand, Dell should have a convincing case that it is the right partner for this.

Enterprise IT shops are also looking to public clouds for a bargain. For rolling out a new enterprise application, cloud vendors such as AWS, Google or Azure, certainly seem like a bargain. No upfront costs to buy server or storage. No wasted time in ordering, testing and provisioning equipment. No complications in planning how to back-up the application or how to scale it if grows quickly. No risk of buying too much and having unused compute and storage resources. And if the new application were to fail or found to be unneeded or redundant, there is no wasted capital for servers and storage that now must be disposed.
As discussed in previous parts, Michael Dell cited a key figure in his keynote, saying he believes public cloud services are 2.5 times more expensive than on premise solutions when considering long term costs, which one would expect means applications with heavy compute, networking and storage characteristics. Over a 3-year period, it is easy to see how public cloud expenses would exceed the cost of buying a server and operating it locally.

However, making a business case centred on long term cost is not the only way that Dell Technologies is tackling the challenge of public cloud migration. One way of fighting the attrition of enterprise customers to the public cloud, is to bring the public cloud on premise. Newly released is the Dell EMC Cloud for Microsoft Azure Stack, an on-premises hybrid cloud platform for delivering infrastructure and platform-as-a-service with a consistent Azure experience on-premises or in the public cloud. Everything you need to run Azure Stack is pre-configured on Dell hardware that resides in your own data centre. The Azure Stack APIs, tools, apps and services remain behind the enterprise firewall, and yet remain consistent with the customer's Azure public cloud experience. For Dell EMC, this is an opportunity to sell more PowerEdge R730XD and S-Series switches. For Microsoft, it is an opportunity to keep the customer running Microsoft server software while perhaps luring some, but not all, of the workloads into the Microsoft data centres.

Moving IT spending to a consumption model

One of Dell Technologies' big announcements is that it will now offer a consumption based model for its IT infrastructure products.  This is its answer to the no-upfront cost lure of public cloud services. In a sense, the public cloud companies are paying to pre-deploy compute/storage/networking equipment in their data centres. They earn this money back as they fill it up with customers. At some point, the money collected exceeds the cost of purchase and from then forward until the day it is obsolete the equipment is delivering profit. With interest rates continuing to remain low, Dell can play this same game, and maybe they do not even need to borrow money and can self-finance the operation. For customers who choose the consumption model, Dell will provide the equipment, as the workloads increase and storage drives get filled, the bill increases. Customers pay only for the storage capacity needed. Importantly, the service promises instant access to additional buffer capacity during spikes driven by the business, with payments adjusting to match usage. If not needed any more, Dell can take back the equipment. Of course, the leasing contract will likely have many conditions and clauses, but certainly this model provides a direct answer for project managers comparing this quarter’s heavy capex spending for a new project with a much leaner opex alternative from Amazon Web Services. The flex pricing is initially available for all Dell EMC storage solutions, but company executives suggested that it could be expended across the IT portfolio.

This consumption model is already being used by other tech companies. There are processor cores that are activated and paid-for on-demand; similarly, Infinera offers a Bandwidth On-demand licensing model where network capacity can be activated in 100 Gbit/s chunks when demand justifies it.

Looking for start-up opportunities

Meanwhile, the new Dell Technologies Capital has just been introduced to manage the company's investment strategy. Specifically, Dell Technologies Capital will invest in start-up companies globally on behalf of all the business units, including Dell, Dell EMC, Pivotal, RSA, SecureWorks, Virtustream and VMware.

Dell Technologies figures that it has more than just capital to invest. Targeted start-up will also benefit from Dell's full resources, OEM and go-to-market relationships and global distribution challenge. While this new investment arm has just been unveiled, it has been running for several years and has already staked out equity positions in more than 70 early-stage start-ups. Targeted areas include storage, software-defined networking, management and orchestration, security, machine learning/artificial intelligence, big data/analytics, cloud, Internet of Things (IoT) and DevOps. Two Dell Technologies Capital portfolio companies were highlighted at Dell EMC World this year:

•   Edico Genome, which created the first bio-IT processor and an end-to-end platform designed to analyse the massive workloads associated with DNA sequencing. The company is delivering a pre-configured, out-of-the- box solution with Dell Technologies that enables the analysis of an entire genome in 22 minutes compared to more than 24 hours using standard software.

•   Graphcore, which developed new technology to deliver massive acceleration for Machine Learning (ML) and Artificial Intelligence (AI) applications. The company's Intelligence Processing Unit (IPU) is the first to be designed specifically for machine intelligence workloads.


All start-ups tend to look for an exit for their early investors, usually in the form of acquisition or an IPO. With Dell Technologies now having many companies under its umbrella, one can expect some of the portfolio investments to become acquisition targets. This is the well-known model of Silicon Valley, with some R&D essentially outsourced and the risk/reward is shared.

Sprint Offers Free Unlimited for One Year to Switchers

In the latest volley in U.S. mobile marketing war, Sprint launched a very aggressive pricing campaign to attract users to join its network. The deal promises "FREE Unlimited data, talk and text for one year" with no annual contract for new Sprint customers currently served by  Verizon, AT&T or T-Mobile. After July 31, 2018, the service is priced at $60/month for the first line. The offer is valid until June 30. Customers must bring their own phones.


https://www.sprint.com/en/shop/offers/free-unlimited.html?ECID=vanity:1yearfree#compare