WHITE PAPER Intel Architecture Intel Solid-State Drives (Intel SSD) IneoQuest* Management Platforms Networking and Communications Video Analytics at the Mobile Edge - Optimizing/Monetizing Viewing Experience IneoQuest* enables real-time video analytics for Intel processors-based smart cell Using the IneoQuest* video analytics system, the potential for total video service visibility and monetization can be realized. Traditionally, video analytics have been collated in data centres by pulling data from multiple sources, and attempting to aggregate and make sense of the data after the event. The increasing need for real-time analytics and the sheer volume of data to be processed are making this data collection and aggregation more complex and costly, and require data centre analysts to be knowledgeable about video in order to make use of all the information.. However, thanks to the convergence of IT technologies on access networks and the advancement of Intel architecture, a majority of analytics can now be collated and preprocessed immediately at the network edge, providing meaningful, actionable data that can be used in real-time. This paper describes IneoQuest* s video probing technology running solely on an open, virtualized smart cell based on an Intel processor. The discussion items include why video intelligence has become a necessity at the edge, how network functions virtualization (NFV) is achieved on smart cells, and the innovations that combine IneoQuest and Intel NFV solutions. To do this, IneoQuest has taken their video analytics solution used in caching and backhaul networks, and worked with Intel to run it on virtualized smart cells, allowing data to be collected right at the edge of the network and providing extensive analytics related to the user behavior and viewing experience. By doing this across all smart cells and harvesting the analytical data centrally using the IneoQuest video analytics system, the potential for total video service visibility (and hence monetization) can be realized. As well as passively monitoring the video sessions passing through the smart cell, the embedded IQ agent can also request video streams to enable pre-caching of popular video content, or to provide a complete video testing service right to the edge of the radio access network (RAN) for service assurance testing or service level agreement (SLA) compliance. By combining Intel architecture with IneoQuest Video Analytics, Intel and IneoQuest are enabling new opportunities to optimize and monetize mobile video.
Video Intelligence at the Edge Explosion of Video Traffic in Mobile Network The Internet is as popular as ever, as evidenced in the growth of global Internet users, which grew over five hundred percent in the past ten years to thirty percent of the total world population. According to Cisco* s latest Visual Networking Index (VNI) 1, global Internet traffic is expected to increase three-fold in the next five years. Watching videos online is the biggest source of Internet traffic today. Moreover, the study forecasts video traffic will make up 69 percent of all Internet traffic in 2017, by which time it would take an individual over 5 million years to watch the amount of video that will cross global IP networks each month. Wireless data traffic continues to grow at an unprecedented rate. Mobile Internet traffic is still booming, with Cisco projecting more than 13-fold growth. The explosive adoption of smartphones and tablets, and the consumption of mobile video are the main contributors to the growth of mobile Internet usage. Cisco reported that in 2012, mobile devices accounted for 10 percent of global Internet traffic. Mobile is taking off, shifting where and when viewers consume video content. Video consumption will grow faster on mobile networks than fixed ones, where mobile video traffic predicted growth is 25-fold. Mobile video is turning out to be one of the few types of mobile content that monetizes reliably and drives premium ad rates for content and service providers, but has shifted the Mobile Network Operator s (MNO) revenue strategy from subscriber acquisition to retention and service differentiation as subscribers are increasingly selecting their service and content providers based on their video experience. Exabytes per Month 12 6 0 96% CAGR 2012-2017 Mobile File Sharing Mobile M2M Mobile Web/Data Mobile Video 2012 2013 2014 2015 2016 2017 Source: Cisco* VNI Mobile Forecast, 2013 < 3.5% < 5.1% < 24.9% < 66.5% Figure 1. Cisco* VNI Showing the Growth of Video Traffic Why Do It at the Edge? No two video providers or operators carrying video for third party providers are the same. They each have different requirements for monitoring and analytics depending on their business models and goals, as video streaming services are delivered over managed and unmanaged networks with best effort characteristics. Operators offering subscriber-based video services will want to know and ensure the quality of viewing experience is good. Collecting information on viewership, most watched times, type of devices, protocols, video assets, and last mile infrastructures quality of service is important in order to optimize the network and delivery, and identify potential problem areas that may require architecture revision. CDN Multiscreen & OTT PGW PCRF HSS Evolved Packet Core (EPC) of Mobile Network Operators that do not offer their own subscriber-based video services and are charging a premium to content providers for carrying their video services will expect a certain level of quality. In this case, video monitoring and analytics are necessary to prove compliance with SLAs that ensure subscribers who pay a premium get a high level of service. Operators that do not carry their own subscriber-based video services and are not charging to carry premium video services on their network will want to monetize their distribution services as video traffic is projected to encompass 90 percent of Internet traffic in 2017. Profiling the video services being delivered through their network provides opportunities for new revenue models. Also differentiating their distribution services from their competitors, operators that have operational and behavioral data at hand can prove their ability to guarantee a level of standard in the delivery of video services. Any of the operators described above should have the ability to diagnose delivery problems within the network. The network edge is an ideal location for validating the quality of the video being delivered to the edge of the RAN. Video analytics at the edge can determine and MME SGW enb HeNB GW (Optional) Figure 2. Simplified Diagram of Mobile Networks 2
pinpoint a host of video delivery issues, allowing operators to understand if the problem is related to congestion in the core network or whether the CDN/ caching server feeding the streams is not responding in a timely fashion. Analytics also allow operators to identify head-end packaging issues with the live or video on demand (VoD) streams vs. infrastructure delivery problems after the head-end. From This... Traditional networking topology Monolithic vertical integrated box TEM proprietary solutions Firewall VPN Intrusion Detection System To This... Networking within s Standard x86 COTS HW Open SDN standard solutions : Firewall : VPN SDN/NFV : Intrusion Detection System As new technology at the edge of the network enables operators to offer services and a user experience that cannot be surpassed in terms of responsiveness and performance, many operators will want to find ways to monetize these opportunities. Edge Cloud Powered by Network Functions Virtualization (NFV) Open, Virtualized Based on an Intel Processor Network Functions Virtualization NFV introduces virtualization technologies into the communications network to create a more intelligent, more agile service infrastructure. According to the Network Functions Virtualization Working Group of the European Telecommunications Standards Institute (ETSI) 2, NFV aims to address problems by leveraging standard IT virtualization technology to consolidate many network equipment types onto industry standard high volume servers, switches and storage, which could be located in datacenters, network nodes and in the end user premises. The key goals of the NFV Working Group are to: Reduce capital expenditures (CAPEX) and operating expenditures (OPEX) of equipment Improve time to market (TTM) by reducing the typical network operator cycle of innovation Figure 3. SDN & NFV Are Driving Architectural Transformation Scale up/down targeted services based on geography or customer sets Enable a wide variety of eco-systems and encourage openness through the use of software solutions All of these objectives are achievable via the use of commercial off-the-shelf (COTS) hardware that can be reprogrammed for multiple functional service nodes that are running on proprietary hardware today. NFV is leveraging virtualization technology and adapting it to the benefit of telecommunications application infrastructure. Figure 3 illustrates how each controller connects to multiple nodes, which can act as a router or a component of smart cell architecture. The controller manages the data paths of the nodes it controls. Above the controllers is the orchestrator, which automates the provisioning, coordination, and management of defined services within a data center or telecom network. This layer also interfaces to external applications and systems. Network applications perform functions such as deep packet inspection (DPI), transcoding, traffic engineering, load balancing, and path optimization. The ultimate goal of NFV is to lead towards SDN and enable applications to determine their data paths based on policies in place within the controllers. Open Virtualized The processing headroom offered by general-purpose processors in small cells has created the opportunity for IT technology to be converged into conventional small calls. Powered by an Intel Atom processor, Intel Core processor, or Intel Xeon processor, together with the Intel Communications Chipset 89xx Series and Intel Solid-State Device (Intel SSD), a smart cell 3 not only offers traffic transportation, but it also allows services to run dynamically as virtual resources. Each small cell is installed with OpenStack* agents such as Nova and Glance, while the images of the services are stored in the cloud. When the service is required, the image will be offered based on a SaaS (Software as ) model, provisioned as virtual resources on the smart cell based on an Intel processor. Added with OpenStack cloud orchestration, smart cells can transform the conventional wireless access network to Cloud-enabled, smart wireless access with NFV. On the other hand, the applications that run on the client devices (e.g., tablet, smartphone, and laptop) can also interact directly with these services, which run as virtual resources. The end user s experience will be greatly improved. 3
OpenStack* Agent Cloud Ochestrator OpenStack Agent OpenStack Agent assurance metrics to analyze how user and infrastructure behavior relates to operational performance. As shown in the diagram, the data from the analytics engine is reported back centrally to the IneoQuest Video Management Platforms for centralized analysis and interfacing to thirdparty provisioning and advertising systems. IQ ASM Intel Architecture IQ ASM Intel Architecture IQ ASM Intel Architecture With the explosive growth of video in mobile networks, monetization of mobile video services is creating an essential need for analytics in terms of the end user viewing experience (reduce churn and validate policy control), behavioral analysis, and enabling advertising revenue. Figure 4. Open, Virtualized Based on an Intel Processor This scalable architecture can potentially lower CAPEX and operational costs for network operators because they can provision just what they need. Then, when new service opportunities arise, they can be easily added into this virtualized, cloud-enabled network architecture. To summarize, smart cells built with generalpurpose processors provide intelligence and storage to the access node, which is managed via OpenStack from the cloud, thus giving equipment providers and mobile network operators the ability to: Efficiently operate, maintain, and manage network resource Speed up functional and service deployments Lower CAPEX by reducing the need for specialized equipment and tool chains Innovate to deliver new and enhanced services and revenue streams (e.g., subscription-based charging model) Virtualized IQDialogue* ASM on an Open, Virtualized Based on an Intel Processor Introduction Working with Intel, IneoQuest has deployed its award-winning video analytics solution on a smart cell platform based on Intel processors, providing operators with extensive real-time behavioral and quality analytics for live and on-demand content. This solution provides the unprecedented ability to analyze mobile video usage behavior in real time and historically, while overlaying video session quality Centralized Video Management & Analytics Video Sessions Backhaul Network By placing IneoQuest technology directly in a smart cell, the analytics gathered at the edge of the mobile network can provide an incredibly accurate and indepth view of how well the infrastructure is delivering video and advertising said Kirk George, Director of Marketing at IneoQuest. This information not only provides a window into how well the content delivery network (CDN) and/ or caching is performing, but also how well the delivery network, client devices, and players are behaving across the different adaptive video delivery protocols such as Apple* HTTP Live Streaming (HLS), Microsoft* Smooth Streaming, MPEG Dynamic Adaptive Streaming over HTTP (DASH), or Adobe* HTTP Dynamic Streaming (HDS). Video Analytics & QoE IQ Inside IQ Inside IQ Inside Figure 5. IneoQuest* Solution at the Mobile Edge 4
Implementation/Solution To date, the cloud-enabled smart cell has successfully offered a virtualized IQDialogue* ASM 4 solution via OpenStack cloud management framework. The ASM images are stored on the cloud. Based on the request of a smart cell customer, these images will be loaded via an OpenStack Glance plug-in onto the smart cell, and then run in a virtual machine via an OpenStack Nova plug-in. Orchestrated by the OpenStack cloud management framework, IQ ASM images will be then offered as SaaS based on the subscription model as requested by customers. Video Analytics at the Edge Adaptive streaming provides for a much closer matching of video delivery to the bandwidth available. Apple s HLS allows better tracking of video to the network bandwidth available, and MPEG DASH has even more potential, especially when combined with the new H265 (High Efficiency Video Coding, HEVC) codec, to effectively double the network capacity for (same quality) video. MPEG DASH and HEVC are both new to mobile and will bring with them a host of issues that need to be diagnosed. Quality assurance is only one phase of monetization. To really maximize the revenue streams in the future, networks will need to be more dynamic to cater to shifting loads and differing service requirements. The video analytics (behavioral and operational) collected allow for better tuning of the networks, new revenue-generating services, improved efficiency, and targeted advertisement insertion. By monitoring traffic at the edge of the mobile infrastructure (and especially if caching is moved to the edge), operators gain true visibility into user and video analytics that enable exceptional analysis and profiling of viewing behavior. A user may stop watching a video after 20 seconds because the visual quality is poor (various possible reasons). IneoQuest UI Video A user may abandon a video stream because it took too long to start the video session (CDN issues, congestion, or cache-load delay). A user may give up watching or call in to complain because the video quality keeps dropping below an acceptable threshold (e.g., sporting event). Thresholds may vary across types of content genres (e.g., fast-moving games or action movies). A user may give up watching an asset because it is looping. Behavioral Analytics Operational Analytics Figure 6. IneoQuest* Solution Implementation Within the Small Cell Alternatively, a user may stop watching because an advertisement drops in, and the user changes to a different asset for the duration of the advertisement, and then comes back to the original asset. This is not a quality issue, but could still be profiled on an aggregate basis for behavioral analysis. By combining the power of behavioral and operational analytics, it is possible to determine the reason users behave the way they do with a much higher degree of confidence something which cannot be done with quality assurance or behavior analysis alone. When deployed within the smart cell, IQDialogue ASM provides both operational and behavioral analytics to ensure a high-quality viewer experience. OpenStack* (Nova, Glance) Other Intel Architecture IQDialogue ASM Virtual Machine () PASSIVE Software Switch Intel ACTIVE WLAN 3G/LTE Radio Smart Phone Tablet Laptop Computer 5
Conclusion There are many reasons why monitoring at the edge is extremely valuable to providers. Without visibility into the adaptive video services at the mobile edge, providers cannot ensure quality and service availability. The key to monetization of video services is to combine quality assurance with analytics with advertisement verification. Intel s platform provides the means to deploy IneoQuest s video monitoring technology that gives operators the insight needed to monetize Adaptive Bitrate Streaming (ABS) video services. For more information about Intel solutions for communications infrastructure, visit www.intel.com/go/commsinfrastructure. For more information about IneoQuest, visit www.ineoquest.com. 1 Cisco* Visual Networking Index (VNI) http://www.cisco.com/en/us/netsol/ns827/networking_solutions_sub_solution.html 2 Network Functions Virtualization Introductory White Paper. ETSI. 22 October 2012. Retrieved 20 June 2013 http://portal.etsi.org/nfv/nfv_white_paper.pdf 3 Intel http://www.intel.com/content/www/us/en/communications/communications-wireless-access.html 4 IneoQuest* IQDialogue* ASM http://www.ineoquest.com/iqdialogue-asm-0 INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WAR- RANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. A Mission Critical Application is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL S PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS. Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked reserved or undefined. Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information. The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order. Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800-548-4725, or go to: http://www.intel.com/design/literature.htm. All information provided related to future Intel products and plans is preliminary and subject to change at any time, without notice. All dates provided are subject to change without notice. Intel may make changes to specifications and product descriptions at any time, without notice. Other vendors are listed by Intel as a convenience to Intel s general customer base, but Intel does not make any representations or warranties whatsoever regarding quality, reliability, functionality, or compatibility of these devices. This list and/or these devices may be subject to change without notice. Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. This document contains information on products in the design phase of development. Copyright 2013 Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Atom, Intel Core and Xeon are trademarks of Intel Corporation in the United States and/or other countries. *Other names and brands may be claimed as the property of others. Printed in USA 0913/MS/SD/PDF Please Recycle 329665-001US