Nokia Networks Nokia Networks telco cloud is on the brink of live deployment
With Nokia Networks multi-stack telco cloud approach, operators have the freedom to choose the cloud platform that matches their needs.
Telco cloud is becoming a reality For many in the telco industry, the concept of the telco cloud is no longer a question of if, but a question of when it will become reality. That s because telco cloud is one of the most frequently mentioned terms in the industry today, creating widespread interest. Many operators are investigating its potential and concluding that telco cloud s benefits are too good to ignore. The attractions of the telco cloud are similar to those in the IT environment, namely lower CAPEX and OPEX and increased business agility. The telco cloud cuts CAPEX through on-demand scalability and commoditization. It enables operators to shift application capacity to match service demand, as well as sharing capacity across and within sites. The use and consolidation of commodity hardware also reduces the required investment. Meanwhile, automation and standardization will drive down OPEX. Operations and maintenance is simplified by using the same hardware and software platform for different network functions, while automated processes are run using cloud stacks, especially when deploying new network elements. For instance, in the core domain an IMS network element in a cloud environment can be installed and commissioned in a couple of minutes. Business agility is a further benefit. Operators will be able to create networks for specific services and/or specific customers quickly and automatically, meaning that a service such as Voice over LTE (VoLTE) could be deployed from scratch in a number of hours. The development and implementation phases are reduced from months to days, while the inherent risks involved are minimized and the operator s business agility is increased. In this brochure we look at how Nokia Networks is developing its Liquid Core solution based on the telco cloud. Telco Cloud: some definitions of common terms Cloud computing separates the business from the infrastructure through automated and programmable data centers using virtualized computing and storage delivered as Infrastructure as a Service (IaaS), Platform as a Service (PaaS) or Software as a Service (SaaS). Software Defined Networking (SDN) separates the control plane from the forwarding plane, making the network programmable via open API s with a centralized control function. Network Functions Virtualization (NFV) leverages IT virtualization technology to separate the network functions from the underlying hardware, enabling the use of non-proprietary, commoditized hardware. NFV is applicable to any data plane packet processing and control plane function in fixed and mobile network infrastructures. Cloud Computing Elasticity & business agility Automated, programmable, SW-defined datacenter Virtualized compute & storage IaaS, PaaS and SaaS* Software Defined Networking Separation of control and user data Centralized control Network abstraction Programmable and open API Network Functions Virtualization* Use of IT virtualization technology Network Functions running on standard IT hardware Higher degree of automation and flexibility *IaaS, PaaS and SaaS: Infrastructure-, Platform- and Software-as-a-Service; NFV- ref. to ETSI ISG NFV Nokia Networks telco cloud is on the brink of live deployment 3
Nokia Networks test programs prove telco cloud readiness In 2012, Nokia Networks ran several proof of concept projects to demonstrate the reliability of core virtualization. These projects provided the technology foundation for core network element software running on top of a virtualized infrastructure. In addition, the projects helped identify telco-specific requirements for cloud computing. While the proof of concept projects showed it is straightforward to implement a prototype Liquid Core solution based on the telco cloud, they also highlighted that existing IT cloud platforms needed to be adapted significantly to meet telco demands. For example, the deployment of core network element software on a virtualized infrastructure needs to be accompanied by new developments for managing its operation and in platform security. Nokia Networks published a white paper, entitled Core network virtualization: a proof-of-concept, giving an overview of the results and this can be found at http://nsn.com/ portfolio/liquidnet/liquidcore. In 2013, Nokia Networks followed up this pioneering work by carrying out testing to verify end-to-end VoLTE deployment readiness in a telco cloud and ran several further trials and live demonstrations. Most recently, Nokia Networks ran a practical demonstration of telco cloud technology at the IFA trade fair in Germany in September 2013 with Germany s first live VoLTE calls being made in a public arena. A multi-stack strategy to match operator preferences Cloud computing is the most rapidly changing environment in IT and the telco industry, although currently two main cloud stacks dominate the market. VMware is one of the pioneers in providing cloud infrastructure with its vcloud Suite comprising vcenter on the virtualization layer and vcloud Director as a Cloud Management System. VMware virtualization and cloud technology are widely deployed in enterprises and in operator data centers. VMware and Nokia Networks are co-operating to address requirements arising from virtual network functions and the Nokia Networks Liquid Core solution based on telco cloud. In parallel, the open source software community is building telco cloud momentum. The OpenStack community is following this approach to provide flexible cloud technology. Nokia Networks supports OpenStack by contributing to the project and associated ecosystems. To enable operators to take advantage of all approaches, Nokia Networks advocates a solution that is independent of hardware and cloud technology. This enables the virtualized core network functions and their management function to support several cloud stacks, beyond the current Openstack and VMware vcloud Suite. As other cloud stack vendors become relevant for the telco industry, this Nokia Networks cloud agnostic approach will ensure they are supported. As well as the cost savings, core virtualization delivers greater flexibility in managing network capacity and allows the easy introduction of new services on the cloud. Nokia Networks believes that the easiest way to manage the complexity of network function virtualization (NFV) architecture is to introduce a three-layered approach: The virtualized infrastructure layer (bottom layer) The application and application management layer (middle layer) The service management layer (top layer) 4 Nokia Networks telco cloud is on the brink of live deployment
Nokia Networks products support a layered Telco Cloud Architecture OSS/BSS Service Orchestrator Service agility and multi-vendor integration Orchestration Federation Deployment Monitoring Operations on Demand (OoD) Elasticity Management Cloud Application Manager Deployment via app. templates NetAct FCAPS* Management Highest operational efficiency via extreme automation Virtualized network functions IaaS API IMS TAS MME S/P-GW Registers Virtual Infrastructure Manager Cloud stacks OpenStack Hypervisor VMware (ESXi) KVM High flexibility via multiple cloud stack support VMware (vcenter, vcloud) Data center Hardware (x86 server) *FCAPS: Fault, Configuration, Accounting, Performance, Security The main components of Nokia Networks three-layer approach and the main goals and benefits Nokia Networks products and solutions principally support the application and application management layer, and the service management layer. By not targeting the virtualized infrastructure market, Nokia Networks can focus fully on supporting all the different virtualized infrastructure flavors and vendors without any conflicts of interest. Nokia Networks fully supports VMware and the Open Stack community and is ready to co-operate with any virtualized infrastructure vendor. Multi-vendor integration of the applications can be done using standardized interfaces between the applications, as in traditional networks today. These same standardized interfaces also enable the applications in the telco cloud to be easily integrated with traditional network elements. A key issue is the end-to-end management of the network, spanning both the traditional part of the network and the telco cloud. The impact on the existing OSS/BSS systems needs to be minimized in order to achieve smooth and cost-effective migration. In addition, vendor-specific application management systems can be integrated with the service management layer. Service management needs to manage a service end-to-end, across traditional network elements and the telco cloud. This includes end-to-end service and resource orchestration as defined by NFV. This approach also minimizes the number of multivendor integration interfaces and allows the telco cloud to smoothly interwork with existing non-virtualized network elements that will continue to exist for many years in hybrid networks and without creating a new silo. Nokia Networks telco cloud is on the brink of live deployment 5
Full testing program for 2013 The 2013 Nokia Networks test program for VoLTE on the Liquid Core solution based on telco cloud used VMware and commercial off-the-shelf IT server hardware. The program tested more than 70 VoLTE functional and telco cloud capability use cases, emulating the range of testing that an operator would typically perform when about to go live with a network. The test program verified the automated deployment and elastic scaling of virtualized network elements, live migration of virtual machines from one server to another, and recovery from hardware failures. The program also carried out several functional VoLTE tests, such as IMS centralized services with Camel homing, enhanced SRVCC with transfer of two parallel calls, offline and online charging, multi-device support, video sharing and emergency call handling. Operations on Demand is a resource management process and daily operation automation tool, providing carefully designed operational workflows which can be run by operators in an automated but controlled way. The Operations on Demand Elasticity Management content pack helps operators to make the best use of its cloud resources by the dynamic allocation of physical resources to cope with decreased or peak traffic loads. Cloud Application Manager provides cloud stack abstraction to any management application running on top. It supports Operations on Demand elasticity management and has been successfully tested with vcloud Director provided by VMware and will be extended to support other technologies, e.g. OpenStack. The test program incorporated the new OSS functionalities, Cloud Application Manager and elasticity management with Operations on Demand. Cloud Domain Management Operation on Demand NetAct Cloud Application Manager OAM i/f IMS inum Open TAS Cloud API Hypervisor i/f Virtualization Infrastructure Cloud Management System (CMS) HW Resource Pool Nokia Networks Telco Cloud laboratory verification reference architecture Most relevant cloud test program showcases Nokia Networks successfully ran five main showcases to prove the capabilities of Operations on Demand and Cloud Application Manager: Showcase One: Automated deployment of a new network element to the cloud In this showcase, Cloud Application Manager deploys various network elements into a VMware-based cloud infrastructure. These complex network elements comprise many virtual machines connected to each other through several network interfaces. Some are also connected to a central storage. Cloud Application Manager reserves all the cloud resources needed, creates the necessary IP configurations, connects the storage, sets up affinity rules and other environmental necessities. The newly deployed elements are preconfigured and are ready within minutes to be connected to the rest of the network and serve traffic. 6 Nokia Networks telco cloud is on the brink of live deployment
Showcases Two and Three: Scale-in and Scale-out These showcases demonstrate the capability of the system to adjust its resource use to the actual traffic needs. Operations on Demand monitors the load on the network elements. When the system becomes overloaded, Operations on Demand initiates a scale-out operation by triggering Cloud Application Manager to deploy additional scaling units of the network elements. During scale-in, when the system loading drops, Operations on Demand starts a process to remove scaling units from the load sharing groups, so the load balancers do not direct new traffic to those units. The system then waits until the ongoing calls/sessions are finished. Once the units are empty, Cloud Application Manager stops the application and the virtual machine, and releases the resources back to the cloud for other applications. Scaling-in and scaling-out operations can be fully automated or configured for semi-automated or manual mode. Showcase Four: Resiliency, system recovery and alarm correlation in NetAct This showcase demonstrates the ability of the network elements to automatically recover from a hardware failure. The loss of a blade carrying several virtual machines was simulated. First, the application level redundancy kicks in and ensures service is not or only minimally affected. The platform recognizes the blade loss and re-deploys the lost units on a new blade. The network element is made fully redundant again. This automated operation happens within minutes, without human interaction. The incident initiates several alarms from the different layers of the system. The virtualization layer sends an alarm about the loss of a host, while applications report the restart of the virtual machines on that given host. NetAct identifies the root cause of these alarms as being the same and correlates the individual alerts into a single alarm. Showcase Five: Virtual Machine live migration The migration showcase demonstrates the system s ability to move a live, running virtual machine from one host computer to another in the virtual infrastructure. During this operation, the virtual machine is able to continuously serve traffic without interruption. In the showcase, the system was 70% loaded, in other words it was running the allocated CPUs at 70% load. Once the migration is initiated, the system creates a copy of the running virtual machine on the target host by synchronizing the memory of the virtual machines and running CPU operations in lock-step mode on the two hosts. As soon as the two virtual machines are fully synchronized, the system switches to the one on the target host and removes the original. Developing multi stack capabilities As well as demonstrating the viability of telco cloud capabilities and a wide range of VoLTE functions, the test program has spawned greater cooperation between Nokia Networks and VMware, resulting in new capabilities for the vcloud Suite for better support of virtualized core network elements running in the cloud. In addition, a sponsorship program allows Nokia Networks to evaluate the capabilities of OpenStack open source software as a cloud management system for virtualized core network elements. By adopting a cloud-agnostic approach, Nokia Networks is well positioned to build on its telco cloud experience and leadership by taking advantage of common off-the-shelf hardware and any vendor cloud stack. In its efforts to further develop carrier-grade cloud technology, Nokia Networks cooperates with major operators, leading virtualization and cloud technology vendors such as VMware and industry forums, members of OpenStack ecosystem and ETSI Network Functions Virtualization Specification (NFV) Industry Specification Group*. Nokia Networks has shown how the complete lifecycle of a network element can be managed remotely via its OSS including initial network element cloud deployment, monitoring of virtualized VoLTE application resources, and elasticity management. Further tests and trials in 2013 and beyond will aim to test the virtualization of other network elements such as the packet core and Home Subscriber Server. In September 2013 at the IFA fair in Berlin, Nokia Networks proved that its Liquid Core solution based on telco cloud is on the brink of live deployment with the live implementation of virtualized VoLTE with a leading global operator. It is expected that the first Nokia Networks virtualized core products will be commercially available from 2014 onwards. * Nokia Networks is an active member of ETSI Network Functions Virtualization (NFV) Industry Specification Group and shares operators objectives to implement network functions in software, and eliminate proprietary hardware components Nokia Networks telco cloud is on the brink of live deployment 7
Nokia Solutions and Networks Oy P.O. Box 1 FI-02022 Finland Visiting address: Karaportti 3 ESPOO Finland Product code C401-01096-B-201410-1-EN Nokia Solutions and Networks 2014 About Nokia Nokia invests in technologies important in a world where billions of devices are connected. We are focused on three businesses: network infrastructure software, hardware and services, which we offer through Nokia Networks; location intelligence, which we provide through HERE; and advanced technology development and licensing, which we pursue through Nokia Technologies. Each of these businesses is a leader in its respective field. Nokia Networks is the world s specialist in mobile broadband. From the first ever call on GSM, to the first call on LTE, we operate at the forefront of each generation of mobile technology. Our global experts invent the new capabilities our customers need in their networks. We provide the world s most efficient mobile networks, the intelligence to maximize the value of those networks, and the services to make it all work seamlessly. www.networks.nokia.com Nokia is a registered trademark of Nokia Corporation. Other product and company names mentioned herein may be trademarks or trade names of their respective owners. networks.nokia.com