Enterprise IPTV Solution Integrated Content Delivery Networks Approach Introduction The Enterprise IPTV Solution enables organizations to produce and broadcast live and on demand video to internal (employees) and external (partners, customers, investors, analysts) audiences over IP based networks. Broadcasts are delivered to auditoriums, conference rooms or desktops, and viewed with media players on desktop computers, or from dedicated Set Top Boxes and multimedia appliances. In essence, it enables organizations to reach dispersed audiences with the same powerful broadcast capabilities that were previously only accessible to the largest media companies. IPTV Viewer based on standard browser and media players Launch Video on Demand or live events from existing web pages or dedicated IPTV web pages Alternate between different formats: video only, presentation only, video plus presentation
Enterprise IPTV Solution 2 In this document we focus on two critical components of an IPTV Solution: Enterprise Content Delivery Networks for the Intranet and Internet Content Delivery Networks for the Internet. We first provide an overview of an Enterprise CDN solution, based the Cisco Application and Content Networking System (ACNS) platform. We then provide an overview an Internet CDN, the AT&T Intelligent Content Delivery Service (icds). Finally, we provide an overview of the Jubilant ERM 2 Platform for IPTV Management, and discuss how ERM 2 integrates with the Cisco ACNS Platform as well as with the AT&T icds service to enable a seamless end-to-end solution for an enterprise. Solution Architecture The Enterprise IPTV Solution is comprised of four key architectural elements, as follows: Presentation Systems and Encoders to incorporate audio, video, and other graphical elements such as Power Point presentations and PC content (spreadsheets, documents, web content) to create multimedia presentations Endpoints to ingest multimedia content, including desktops, Set Top Boxes and portable media devices Distribution Systems to efficiently and securely distribute multimedia content throughout the enterprise and on the Internet. This includes systems and technologies such as Enterprise Content Delivery Networks, Internet Content Delivery Networks, Multicast and QoS IPTV Management System coordinates and controls the Presentation Systems, Endpoints and distribution systems. It also integrates with existing enterprise applications such as portals, Identity Management systems and search engines Enterprise IPTV Architecture Enterprise Applications Portals Search Engines Identity Management Jubilant ERM 2 Platform IPTV Management System Enterprise Network Infrastructure IP Multicasting, QoS, Content Routing Enterprise Content Delivery Network Internet Content Delivery Network Enterprise Communications Video Conferencing Multimedia Systems Encoders Capture Stations Set Top Boxes The Big Picture: The Enterprise IPTV Solution is comprised of an IPTV Management System and Multimedia Systems. The IPTV Management System integrates with existing Enterprise Applications, Communications Systems and Network Infrastructure Enterprise IPTV High Level Design Regional Office Home Office, Customers, Partner Regional Office AT&T icds (Internet) AT&T MPLS (Intranet) The Enterprise IPTV Management System is the central point of management for the Enterprise IPTV Solution and provides three key value propositions. First, it manages and coordinates all the systems that comprise the solution to enable a streamlined, automated workflow. Secondly, it provides a full suite of management capabilities for producers and administrators, to enable the scheduling, presentation, distribution, and publishing of multimedia content, with minimal support from corporate IT staff. Third, it integrates with existing enterprise applications and network infrastructure to enhance and optimize IPTV services in the enterprise. Enterprise Content Delivery Network Enterprise Content Delivery Networks (ecdn) can significantly decrease the impact that multimedia content such as audio and video can have on the enterprise environment. In essence, content is accessed by users from local edge or branch site caches so that users do not need to access content directly from the origin server that may be on the other side of a WAN cloud. Most ecdn platforms also provide proxy cache capabilities with added features to ensure content freshness. In particular, these platforms exploit features of the HTTP protocol in which content is conditionally retrieved based on how recently the content was modified, or whether the content was modified at all. Other features of an ecdn can include the following: Support for live and on demand media Full application proxy caching for HTTP, HTTPS and FTP RTSP protocol proxy and server functionality Jubilant ERM 2 IPTV Management System AT&T Data Center or Customer Premises Production Studio Customer Data Center Application-layer multicasting for live video optimization Encoder and Camera Enterprise Applications (Portal Identity Management Systems, Search Engines) Performance improvements to web-based enterprise applications
Enterprise IPTV Solution 3 Cisco Application and Content Networking Solution (ACNS) is an ecdn platform that can be utilized by the AT&T Enterprise IPTV Solution. Cisco ACNS supports a number of mission-critical network services using a common infrastructure of network-integrated and dedicated appliances called Wide Area Application Engines (). devices are packaged as standalone network appliances or router-integrated network modules (NME-). Running Cisco ACNS software, devices allow organizations to accelerate the delivery of rich media and multimedia applications, particularly at the edge of the network in branch offices, where bandwidth tends to be scarce. By serving content closer to end users, Cisco ACNS allows organizations to reduce the bandwidth required for web-enabled applications and restrict bandwidth utilization to sanctioned business information and applications. In this section, we describe the high level features and functions of the Cisco ACNS Platform, looking first at how Cisco ACNS supports live video streaming, and then on how Cisco ACNS supports Video on Demand. Live Video Streaming A publicly announced live streaming event is the single most taxing event on a non-multicast enabled network and on the origin servers. In a live streaming event that does not take advantage of network multicast or application-level stream splitting, a separate stream is created to the originating video server from every single client. If the stream encoding rate is reasonably high at 300 Kbps and there are 10,000 users, video servers will have to deliver 3 Gbps of video streaming traffic. The data centre deployment would require a large number of video servers to support the event, ignoring the fact that the rest of the network will likely encounter congestion before the video servers see all the requests. The strain is compounded if there is a WAN between the clients and the video servers. Live Stream Without ACNS A separate stream for each client traverses the WAN Aggregate data traffic for all clients must be less than the WAN bandwidth Unicast WAN In the case of unicast stream splitting, clients get to view the broadcast when it occurs. The intermediate s cause minimal delay compared to the delay introduced by the encoder, encoder buffer and media player buffer. With stream splitting, the customer does not have to upgrade network infrastructure to enable IP multicast to achieve reliable, high quality live streaming. Live Stream Unicast Stream Splitting Overcomes WAN Bandwidth Bottleneck Requires Larger CEs Live Unicast Video Multiple Unicase Streams (One per user) Splitters Unicast WAN DNS CDM Video Server Finally, the customer does not have to create a scheduled multicast event because WCCP-enabled s automatically split the stream. In all circumstances, video quality is limited to the available WAN bandwidth. Customers require a full duplex switched Ethernet LAN infrastructure at the remote sites because each client pulls an additional unicast stream. Hybrid Unicast to Multicast Streaming Back channel or unicast-to-multicast streaming is similar to stream splitting except that the final leg of the stream is converted to a multicast to minimize the bandwidth impact on the network and the load on the. In many cases the last leg will be co-located with the client in which case the acquires a single unicast stream over the WAN and converts it to a multicast in the LAN. If a customer has a multicast enabled WAN then they can use a to publish the multicast in the data centre to stream the multicast to all the multicast enabled clients over the WAN. The multicast enabled network routers and switches replicate/copy the streams (packets) where necessary so that all clients can view the video event. Live Stream Hybrid Unicast to Multicast All Requests DNS Video Server Unicast Stream Splitting Live unicast stream splitting is intended for instances where there is no network multicast enabled and/or there is a desire to provide live streaming for ad hoc, unplanned events. When there is a Cisco with each WAN router, unicast stream splitting supports live streams that originate on the non-multicast enabled Internet or within the corporate intranet. Multicast enabled LAN only CE scales to many simultaneous programs Live Unicast Stream Single Multicast stream Replicated by Network DNS Unicast WAN CDM Video
Enterprise IPTV Solution 4 Video on Demand Streaming The challenge with VoD has traditionally been WAN bandwidth. However, in the case of a that is local to the client requesting content, this WAN bottleneck can effectively be removed. VoD content is generally comprised of files generated from a live event. It takes the form of a Microsoft Windows Media file, a Real media file, or other standard video formats (MPEG-4, H.264 etc.) Video on Demand Without ACNS Separate stream for each client across the WAN Aggregate data traffic of all clients must be less than the WAN bandwidth Internet or WAN for multiple clients to simultaneously view the video. Caching is the only option if the organization that controls the does not have access to the video files before a client request is made. The Cisco ACNS solution is capable of caching video files in a variety of formats such as Microsoft Windows Media, Real Networks, and QuickTime. A primary advantage to caching is that there is no administration required. The disadvantage is that the first user can experience poor quality if there is network congestion. Pre-Positioning for Video on Demand In the case where the recorded video is owned or can be acquired, it may be pre-positioned by Cisco ACNS using administrator defined bandwidth distribution controls, or by utilizing third party systems that integrate with Cisco ACNS API s and have intelligent and automated content distribution capabilities, such as the Jubilant ERM 2 Platform, as described in this document. Video on Demand Pre-Positioned Video First Request Subsequent Requests CR CDM Video Server The video file may be pushed (pre-positioned) to a local using the CDN file distribution capabilities (aka content pre-positioning) or pulled as a result of a client request. These two mechanisms are described in further detail below. Regardless of how the video file reaches the, the file will be delivered locally across the LAN once it is in the cache. The video storage requirements can generally be calculated as follows. If the stream encoding rate is 300Kbps, then an hour of video will consume 300,000 bits/second * 3,600 seconds/hour * 8 bits/byte or 135 MB. Based on these estimates, a with 40 GB capacity will hold about 296 hours of video. Pull Caching for Video on Demand Pull caching allows the to dynamically learn what content should be populated in the cache by waiting for a user to request it. Caching is recommended if the encoded video (or streaming rate) is smaller than the available WAN bandwidth, but there is insufficient bandwidth Video on Demand Pull Caching Streamlined bandwidth is greater than WAN bandwidth Pre-positioning All Requests Internet Content Delivery Network Internet or WAN DNS CDM Video Server Overview AT&T s Intelligent Content Distribution Service (ICDS) can significantly improve your website s capacity, reliability and performance. By serving your traffic through our worldwide network of high-speed edge servers, your web content and media streams reach users faster and more reliably than from your servers alone. ICDS can reduce your server farm bandwidth and horsepower requirements, and enjoys the advantage of AT&T s extraordinary network capacity and impressive peering capabilities. ICDS Distributed system increasing net performance, reliability and economics Origin Site Periodic Bottleneck Streamlined bandwidth must be less than WAN bandwidth Internet or WAN User Access Network Internet Content First Request Subsequent Requests DNS CDM Video Server
Enterprise IPTV Solution 5 AT&T can combine content delivery, broadband/dial or dedicated access and advanced hosting facilities with professional engineering support services to provide a single vendor end-to-end solution. AT&T utilizes its own IP backbone, making AT&T uniquely positioned to find the best way to manage the traffic on content distribution. AT&T Intelligent Content Distribution Service is a facility-based Content Delivery Network that replicates information across the Internet. With ICDS, end users can get the fastest possible download since content is placed on multiple servers located closer to your web viewers. Benefits Extremely large burst capacity Fault-tolerant node infrastructure reduced single point of failure vulnerabilities Internet connectivity via the AT&T OC48/OC192 backbone network and to more than 140 additional ISPs Global and Local Load Balancing Server clusters Support for HTTP 1.1 Cache Control Headers Features Intelligent Domain Name System Enables ICDS to route content to the most optimal edge node Highly Available Distribution Architecture Nodes are placed throughout the AT&T network, are connected to each other by high-capacity and redundant links AT&T BusinessDirect Portal XML-based Cache Control Preload or purge the ICDS network cache-specific files Capabilities AT&T s Intelligent Content Distribution Service offers Caching services to our clients. Cache servers for the caching service operate in the reverse proxy mode and appear as real web servers to end-user s browsers. Cache servers pull cacheable objects from origin servers on a needed base; non-cacheable objects are bypassed and sent to origin servers directly. This service is further divided into two offers. Web Acceleration Web Acceleration is the basic offer of caching service. For this offer, customers host content on their own origin servers. ICDS accelerates all or a portion of HTTP/HTTPS objects for a customer s web site. Typical objects accelerated by Web Acceleration are HTML files and image files. Large File Download (LFD) LFD (or File Download or efile Download) is a special version of caching service. Web sites identified as LFD type typically contain large size objects such as downloadable software or QuickTime files. For this offer, customers may host content on their own origin servers or use network origin servers and storage provided by ICDS. Additional Services SSL Caching The Secure Socket Layer (SSL) caching feature will allow customers to use ICDS to accelerate static content of their secure web sites Cookie Filtering This feature will allow each caching site to specify whether ICDS should ignore cookies or not Support of Referral URLs This feature will allow customers to specify whether ICDS should serve a site s objects to end-users or not based on the referral URL received in a request Intelligent-Site (I-Site) Management I-Site Management utilizes an Intelligent Domain Name System (idns) developed by AT&T labs research group to dynamically throttle content requests between a customer s origin servers and ICDS cache servers. I-Site Management is ideal for web sites that do not require to be accelerated by ICDS all the time. I-Site Management is offered as an add-on for the caching service described above. I-Site Management includes Capacity On Demand (COD) and Slim Mirror features Capacity On Demand (COD) COD allows a customer to specify a load cut-over range within which idns will gradually divert content requests away from the customer s web servers and into ICDS cache servers. COD is useful for customers that want to use their own origin servers for content requests if their servers can easily handle the load; but when the load on their servers increase, customers can depend on ICDS to manage the load and gradually take over subsequent requests Slim Mirror Slim Mirror allows customers to use ICDS as a backup/failover service. idns will periodically check whether a customer s web site is available. If the web site does not respond in an expected manner, idns will switch end-user requests to ICDS cache servers, which in turn will fetch content from a separate backup web server provided by the customer. This backup web server may contain a scaled down version of the original web site and reside on a less powerful hardware platform ICDS also utilizes idns to perform Proximity Routing, an advanced Global Server Load Balancing (GSLB), for caching and streaming services. Behind each cache node, local Server Load Balancing (SLB) is used to balance content requests within a node.
Enterprise IPTV Solution 6 Deployment ICDS Methodology ICDS Node Management System F/W APP CMS RPT DB Web Acceleration Node Proxy Cache Management Servers Application Servers Portals Database Servers Reports Servers Operations Portal (on App. Servers) Customer Portal (on App. Servers and Proxy) Firewall AT&T IP Network Log File Servers Intelligent DNS 4 Locations VoD Storage Media Servers Proxy Servers Provisioning Servers Gigabit Filter Cache Servers L2 Switch Gigabit Filter Streaming Cache (5) for Live Webcasting Layer 4-7 Switch Streaming Cache (2) For VoD Publishing Server Complex (SPPOC) Proxy Cache Network Origin Servers On Demand Storage Provisioning and FTP Server Deployment Methodology: Regional and Global Content Propagation The ICDS service architecture comprises many systems and functions. Within the ICDS caching service, content is propagated from customer Origin Servers to AT&T ICDS cache servers when needed to fulfill an end-user content need. The figure above shows the relationship between each system. Sourcing Content ICDS uses a pull method based on end user requests to get the content from the source. If the requested content is not present in a cache server, the cache server will pull the content from the customer s origin server. This can occur either directly between all cache servers and the origin server, or the cache servers can be logically arranged in a parent-child hierarchy, as shown below, to reduce the load on the origin server. If one child cache server makes a request to a parent server that already has the content requested, the parent will return the content to the child. ICDS Request Flow Distribution of Content When an end-user tries to use a web browser to access an object that is served by ICDS, the following steps are carried out: 1. The web browser parses the URL and issues a DNS lookup to the local DNS server to resolve the domain name embedded in the URL. 2. The local DNS server processes the DNS lookup request by first checking its local DNS cache. If the local DNS cache can satisfy the DNS lookup request, the local DNS server returns the corresponding IP address back to the web browser. Note: the DNS Time-To-Live (TTL) value is usually set to very short like 2 minutes; this enforces local DNS servers re-query periodically. Proceed to Step 6. 3. If the local DNS server cannot fulfill the DNS lookup request, it follows DNS referral rules to route the DNS lookup request. At the end, either the DNS server in the layer 4 switch or idns gets the request. 4A. The layer 4 switch uses Global Server Load Balancing (GSLB) scheme to determine which cache node should serve the content request and return the Virtual IP (VIP) address back to the local DNS server. DNS Request for.nasa.gov Closest Node s IP http request/response for page End User (near Denver) Child Nodes Atlanta Austin Cambridge Denver Detroit Lisle NYC, NY idns Hawthorn, CA Cache Nodes 1 Parent Nodes Denver Dallas Origin Servers (Ashburn, VA) Switch Web Servers Or 4B. idns uses Proximity Routing scheme to determine the cache node closest to the local DNS server and returns the VIP address back to the local DNS server. Note: idns is not available for SSL caching option. 5. The local DNS server returns the (virtual) IP address back to the web browser. 6. The web browser attempts to establish connection with the returned IP address. 7. The layer 4 switch uses local Server Load Balancing (SLB) to route the content request to a cache server in the cache node. 8. The web browser issues an HTTP (for Web Acceleration and LFD) or HTTPS (for SSL caching option) request with the cache server. Seattle 9. If the requested object is cached by the cache server and still valid Request for content only if cache doesn t currently have the object 1 Nodes contain from 2 to 6 servers each
Enterprise IPTV Solution 7 (e.g., not expired), the cache server returns the object back to the web browser. Proceed to Step 13. 10. If the requested object is not cached by the cache server or is determined invalid (e.g., expired), the cache server looks up its configuration to find the origin server and initiates an HTTP (or HTTPS) connection to the origin server. Note: for hierarchical caching, the origin server may be another cache server; if the object is not cached by the parent cache server, it will then initiate a connection to the origin server. 11. The cache server sends the URL requested by the web browser to the origin server. 12. The origin server returns the object back to the cache server. If the object is cacheable, the cache server makes a copy of the object in its local storage. 13. The cache server returns the object back to the web browser. Default method for ICDS is filling of the cache based on end user request (Request driven). But ICDS also provides tools to customers if customer desires to pre-populate the content. Pre-population can be either through the web GUI on Customer Portal or through the use of the XML interface. Fault Tolerance Performance, high availability, and reliability are prime motivating factors in the design of the ICDS network. Key points in the design and operation of ICDS that achieve these goals are explained below. Redundancy in Caching and Streaming Architecture The ICDS network contains redundant architecture for caching and streaming service. The ICDS network is based on multiple geographically distributed caching nodes, each of which contains multiple high-performance high-bandwidth caching appliances. Proximity-Based Load Distribution ICDS requests are directed to a node that has a good probability of being close (in terms on network latency) to the requesting client. Various techniques are employed to achieve this result, and refinements are made on an ongoing basis to the tables and algorithms underlying this feature. Should a nearby node fail or become heavily loaded, the system will adapt by rerouting requests to another favorable node. Traffic-Based Load Distribution ICDS compliments its proximity-based balancing with a traffic-based model when a given node becomes heavily loaded. This assures that the proximity-based routing does not overwhelm any given node just because clients happen to be closest to that node. Traffic-based load balancing distributes requests in a fair manner across the less-loaded nodes, always striving to keep the loads evenly distributed. Local Server Load Balancing Within any given node, multiple servers are available to handle requests. Failure of one or more servers is automatically detected and such servers are immediately removed from the node s server rotation list. These servers are brought back online only after Operations has isolated and corrected the problem. Redundancy in Origin Servers for Caching Customers may employ multiple origin servers for the same web site for caching. ICDS caching can be provisioned with multiple origin server IP. Requests will be balanced across the multiple servers and failed servers will be skipped and periodically retried. Redundancy in Local Management Servers for Caching Each of the 16 domestic U.S. caching nodes contain two Local Management Servers (LMS) for operations support. Both LMS will share processing of raw log files from cache servers. Not only more logs can be processed at each node, but a node can continue function even if one LMS server is down. Redundancy in Large File Download Customers can employ multiple origin servers similar to the redundancy in the Caching service. For Network-based origin servers for Large File Download (LFD), ICDS provides two geographically diverse locations with two origin servers in each location. Customers can upload the content to one location, and the content is replicated to the other location within 20 minutes 90% of the time. DNS Query Distribution ICDS supports DNS traffic as part of its load-balancing feature. DNS A record queries for customer ICDS domain names are resolved within the ICDS nodes, the responses chosen so as to direct HTTP traffic to the best node, as described above. To assure continual DNS availability, multiple DNS servers are deployed. Most nodes contain one such server. Should a DNS server fail, the public DNS system will simply select from among the remaining servers. One or all of the cache servers within a node can fail without loss of service. Should an entire node fail, requests that would have been directed to it will generally be directed instead to the next most favorable node in terms of network latency. Likewise, the failure of multiple nodes will not interrupt service as long as at least one node remains to serve the customer. The exact set of nodes any given customer is provisioned into depends on that customer s Capacity Type Group (CTG). CTGs are selected based on the feature set required by the customer. Scalability and Monitoring Resource Availability AT&T uses the award winning igems infrastructure to monitor all hardware and software supporting the ICDS service. Alarms are generated based on the thresholds that are set and the NOC responds proactively to the alarms before the situation becomes critical. There are multiple levels of alarms (normal, warning, major, critical) for CPU, bandwidth, storage, as well as many other parameters. AT&T ICDS service also has a capacity planning process; The Life Cycle Management organization monitors the service capacity using proprietary trending tools. These tools produce forecasting statistics that predetermine when additional capacity needs to be deployed. IPTV Management System The Enterprise IPTV Management System is the central point of management in the Enterprise IPTV Solution and provides three key value propositions. First, it manages and coordinates all the systems that comprise the solution to enable a streamlined, automated workflow. Secondly, it provides a full suite of management capabilities for producers and administrators, to enable the scheduling,
Enterprise IPTV Solution 8 Focus on the Enterprise IPTV Managment System Features Enterprise Applications Portals/Collaboration Learning Management Search Engines Enterprise IPTV Management System Encoder Management Publishing Identity Management Systems Management Content Management Event Management Enterprise Network Infrastructure and Transport IP Multicasting, QoS, Content Routing, VPN Routing and Switching User Management Access Control & DRM Enterprise Communications Voice over IP Web Conferencing Video Conferencing Distribution Management Set Top Box Management Application Acceleration Enterprise/Internet Content Delivery Network, MPLS Media Indexing System Monitoring Performance Monitoring Reporting Presentation Systems Encoders Endpoints Desktops Set Top Boxes Multimedia Appliances presentation, distribution and publishing of multimedia content, with minimal support from corporate IT staff. Third, it integrates with existing enterprise applications and network infrastructure to enhance and optimize IPTV services in the enterprise. Problems with IPTV in the Enterprise In the past, producers had to interface with a variety of corporate IT staff responsible for enterprise applications (portal, security) as well as network infrastructure in order to plan and execute IPTV events. Furthermore, producers had to coordinate and manage a myriad of multimedia systems, including encoders, streaming servers, presentation systems and Set Top Box Systems. Before an event, producers needed to interact with IT staff to publish and announce upcoming events on corporate portals, and also interact with network engineering teams to plan for the distribution of multimedia traffic in the Internet as well as on the corporate network. During an event, producers relied on support from IT staff to monitor the distribution of live multimedia traffic throughout the enterprise network, and after an event, producers had to interact with IT staff to publish Video on Demand content onto corporate portals and plan for the distribution of Video on Demand files on the Internet and throughout the corporate network. Error-Prone Manual Processes and Communications These manual processes lead to errors, which in turn lead to unreliable services. The manual processes also incur significant hidden costs in the number of IT personnel involved in each broadcast. Solution: IPTV Management System The IPTV Management System eliminates these manual processes and enables a streamlined, automated workflow. It provides a full suite of management capabilities for producers and administrators, to enable the publishing, presentation, and distribution of multimedia content, without the need for support from corporate IT personnel. Enable Self-Service Processes Self-service model for Producers without the need to interface with Corporate IT or Network Infrastructure Corporate IT Production and Media Services Enterprise Applications and Communication Systems Portals/Collaboration Learning Management Search Engine Identity Management Systems Managements Enterprise IPTV Managment System Streamlined workflow and central point of management for Enterprise IPTV Services Voice over IP Web Conferencing Video Conferencing Multimedia Systems Presentation Systems Encoders Set Top Boxes Multimedia Appliances Corporate IT Production and Media Services Network Infrastructure Enterprise Applications Portals/Collaboration Learning Management Search Engine Identity Management Systems Managements Presentation Systems Enterprise Communications Voice over IP Web Conferencing Video Conferencing Encoders and Streaming Servers Set Top Boxes and Multimedia Appliances Enterprise Network Infrastructure and Transport IP Multicasting, QoS, Content Routing, VPN Routing and Switching Application Acceleration Enterprise/Internet Content Delivery Network, MPLS Network Infrastructure Enterprise Network Infrastructure and Transport IP Multicasting, QoS, Content Routing, VPN Routing and Switching Application Acceleration Enterprise/Internet Content Delivery Network, MPLS Workflow Pre-Production Producers provide scheduling information for events to the IPTV Management System, which publishes and announces upcoming events on corporate portals, provisions network infrastructure such as Enterprise Content Delivery Networks and Internet Content Delivery Networks, and provisions multimedia systems such as encoders and Set Top Boxes. Producers also specify who can view what event during this phase to enable access control for live events. Problems with IPTV in the enterprise
Enterprise IPTV Solution 9 Production The IPTV Management System controls and monitors multimedia systems (encoders, Set Top Boxes), monitors the performance of the video streams, and gathers reporting information for each event. The IPTV Management System also enforces access control for users accessing content which was specified by producers prior to an event. Post-Production The IPTV Management System archives, processes and indexes recorded events, publishes content in a searchable Video on Demand library and also posts Video on Demand content onto portal pages. The IPTV Management System can also pre-position content on Enterprise or Internet Content Delivery Networks. This streamlined workflow enables reliable IPTV services and greatly reduces the hidden costs associated with the number of technical support personnel involved in each broadcast. In the remainder of this section, we provide an overview of the Jubilant ERM 2 Platform for Enterprise Rich Media Management, which serves as the foundation of the Enterprise IPTV Solution. We first summarize the key capabilities provided by the platform, and then summarize the integration points that are possible between the platform and other enterprise applications and network infrastructure. Capabilities The following is a summary of the capabilities provided by ERM 2 : Publishing. Announce upcoming events in a centralized Program Guide or portals and post Video on Demand content in existing web pages Multimedia Search. Provide a searchable archive of Video on Demand content Event Management. Schedule live and rebroadcast events Encoder and Presentation System Management. Configure, control and monitor remote encoders and Presentation Systems to synchronize audio, video and other graphical content e.g. slides User Management, Access Control and Digital Rights Management. Encrypt multimedia content and enforce access control Content Management. Manage recorded events in a virtual folder hierarchy Media Indexing. Automatically index recorded events to enable search based on phrases, keywords, acronyms, proper names Distribution Management. Configure and provision both Internet and Enterprise Content Delivery Networks for the distribution of live and on demand video content Set Top Box Management. Configure and control remote Set Top Boxes, publish program guides on Set Top Boxes and enforce conditional access from Set Top Boxes Reporting. Provide information about who viewed what, where and when System Monitoring. Monitor encoders, Presentation Systems, application servers, Set Top Boxes and Content Delivery Networks Enterprise Integration ERM 2 integrates with existing enterprise applications and network infrastructure to enhance and optimize the performance of IPTV services in the enterprise. The following is a summary of the key integration points. Identity Management Systems enable group based access control Portals provide a launch point for Program Guides and Video on Demand Libraries Search Engines enable multimedia search capabilities for Video on Demand Libraries Enterprise Content Delivery Network on the customer s network can be utilized to optimize WAN bandwidth Internet Content Delivery Networks enable the reliable and secure delivery of high quality video on the Internet Integration with Enterprise Content Delivery Networks ERM 2 enables producers to manage Enterprise Content Delivery Networks (ecdn) such as the Cisco Application and Content Networking System (ACNS), for the efficient and optimized distribution of multimedia content in the enterprise. ERM 2 integrates with the Cisco ACNS Platform to enable two key benefits. First, ERM 2 enables the distribution of Video on Demand files via the ecdn. Video on Demand content is centrally managed from one of the streaming servers, but local copies can be pre-positioned to the Cisco ACNS edge cache devices, also known as Content Engines (CE). Users accessing content are automatically redirected to stream the video from their local CE, reducing the load on the WAN as users across the enterprise stream Video on Demand to their desktops. Second, ERM 2 automates the configuration and provisioning of the ecdn for live broadcasts. As broadcasts are scheduled, producers specify which subset of edge devices should receive each broadcast i.e. based on region, business unit etc. Based on this information, ERM 2 configures the ecdn for broadcasts, without requiring producers to communicate with network administrators to manually configure the ecdn for each broadcast. This capability to preposition Video on Demand content as well as configure and provision the ecdn for live broadcasts is enabled by the API provided by the Cisco ACNS Platform. Integration with Internet Content Delivery Networks ERM 2 enables producers to manage Internet Content Delivery Networks such as the AT&T Intelligent Content Delivery Service (icds) for the efficient and optimized distribution of multimedia content in the enterprise. ERM 2 integrates with AT&T icds to enable two key benefits. First, ERM 2 enables the distribution of Video on Demand files via the ecdn. Video on Demand content is centrally managed from one of the streaming servers, but content can be cached on AT&T icds. Users accessing content are automatically redirected to stream the video from the icds caching servers instead of from the customer s streaming servers, ensuring a high quality experience for users, and preventing performance bottlenecks for customers. This capability to
Enterprise IPTV Solution 10 preposition Video on Demand content is enabled by utilizing the XML API provided by AT&T icds to control the pre-positioning and purging of media files. Second, ERM 2 automates the configuration of the AT&T icds service for broadcasts. As broadcasts are scheduled, producers specify whether a particular event should be available on the Internet. Based on this information, ERM 2 configures the AT&T icds for broadcasts, without requiring producers to communicate with network administrators to manually configure AT&T icds for each broadcast. This paper was written in cooperation with Cisco Systems, Inc. and Jubliant Technologies, Inc. Cisco Systems, Inc. http://.cisco.com/go/acns Jubilant Technologies, Inc. http://.jubilant.biz For more information contact your AT&T Representative or visit us at.att.com/business. 01/29/08 AB-1229 2008 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property.