GLOBAL NETWORK OPERATORS, FROM SEA TO CLOUD

Transcription

1 GLOBAL NETWORK OPERATORS, FROM SEA TO CLOUD Introduction Cloud computing is the latest technology trend that has sparked interest across a wide variety of industries. In the midst of an increasingly competitive environment, corporations must constantly improve the quality of their IT services, even while the number of users and applications is increasing steadily. Traditionally, corporations addressed their growing IT requirements by investing additional capital in equipment; however, IT departments are also under considerable pressure to reduce capital and operating expenses to remain competitive as global competition intensifies. Thanks to recent virtualization advances in the areas of storage, compute, and networking, outsourcing IT infrastructure is becoming an effective option available to corporations of all sizes. Enterprise-class infrastructure is moving into the cloud, enabling on-demand services with high levels of performance, security, and manageability. Global network operators, characterized by owning and operating interconnected terrestrial and submarine networks, are pursuing three key strategies in their quest for new streams of revenue available with cloud services. > Networking Providing network services with global reach, focused on guaranteed Service Level Agreements (SLAs) necessary to hide the growing distance between users and data centers > Partnering Aligning with cloud service providers to access their data center facilities and expertise in IT hardware and software management and deliver bundled cloud-plusnetwork connectivity services > Vertical Integration Owning, operating, and growing their own geographically dispersed data centers via organic growth and/or through corporate acquisitions for a complete cloud offering Regardless of the chosen business approach for offering differentiated cloud services, one thing remains constant: The network connecting users to their data is a key competitive advantage for global network operators. The cloud network, which connects users to their data via overland and undersea networks, must perform at the same level, or better, than that to which users have become accustomed when the data center is collocated within the same building or campus. This means that network performance must be precisely architected and operated from end to end, both overland and undersea, which places global network operators contemplating the introduction of cloud services in an enviable competitive position. According to a recent report 1 on the Reliability of Global Undersea Communication Cable Infrastructure (ROGUCCI), nearly 100 percent of the world s inter-continental electronic communications traffic is carried by the undersea cable infrastructure. This means the undersea jugular veins of intercontinental communication are critical components when designing seamless end-to-end global networks optimized for cloud services. In addition, any physical damage to the submarine cables that stitch together the global network could compromise the global cloud network, and thus must get the network protection it deserves network resilience is now a prime requirement. This paper offers an overview of cloud computing and how it will affect the ways global networks should be architected. Cloud services place unique performance requirements on the network where its services will reside, thus making SLAs a prime concern. Cloud Evolution Cloud computing, a strategic option available to IT departments to right-size their capacities and optimize associated costs, is rightly receiving a great deal of attention. However, one adverse side effect of this attention is that there is confusion as to exactly what cloud computing actually means. The United States National Institute for Standards and Technology (NIST) defined cloud computing as follows. Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model is composed 1 Proceeding of the ROGUCCI Study and Global Summit, Issue 1, Revision 1, 2010, IEEE Communications Society W Whitepaper

2 of five essential characteristics, three service models, and four deployment models. According to the NIST, cloud computing must also meet the following essential characteristics. On-Demand Self-Service Consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service provider. Broad Network Access Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations) Resource Pooling The provider s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources, but may be able to specify location at a higher level of abstraction (e.g. country, state, or data center). Examples of resources include storage, processing, memory, and network bandwidth. Rapid Elasticity Capabilities are elastically provisioned and released, in some cases automatically, to scale rapidly outward and inward commensurate with demand. To consumers, the capabilities available for provisioning often appear to be unlimited and can be appropriated in any quantity at any time. Measured Service Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (such as storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service. There are three service models associated with cloud computing, which have broad consensus across the industry. Software-as-a-Service (SaaS) Software applications that are designed for end-users and delivered over the Web, accessed by a thin client such as a Web browser. SaaS examples include Web-based , Customer Relationship Management (CRM), Enterprise Resource Planning (ERP), and Human Resources Management (HRM) software applications, to name a few. Industry-as-a-Service (IaaS) The hardware and software that powers the cloud, including the storage arrays, application servers, firewalls, networks, and operating systems. Platform-as-a-Service (PaaS) A set of tools and services designed to facilitate application coding and deployment. Finally, there are four generally accepted cloud deployment models, which IDG Communications defines as: Community Cloud Shares infrastructure between several organizations from a specific community with common concerns (such as security, compliance, and/or jurisdiction), whether managed internally or by a third party and hosted internally or externally. Public Cloud The cloud infrastructure is provisioned by the cloud provider for open use by the general public. It may be owned, managed, and operated by a business, academic, or government organization, or some combination. Private Cloud Infrastructure provisioned solely for a single organization, whether managed internally or by a third party and hosted internally or externally. Hybrid Cloud A composition of two or more clouds (private, community, or public) that remain unique entities but are bound together, offering the benefits of multiple deployment models. It can also be defined as multiple cloud systems connected in a way that allows programs and data to be moved easily from one deployment system to another. As cloud services continue to evolve and expand in scope, the enterprise data center infrastructure must be seamlessly integrated into the cloud. Most SaaS offerings today, such as common services, are not mission-critical applications, so performance requirements placed upon public cloud infrastructures are typically modest. However, as mission-critical services and applications are migrated into the cloud, a new high-performance architecture incorporating multiple geographically distributed data centers is required to improve service resilience, resource efficiency, and application execution. Ciena refers to this new network architecture as the Data Center Without Walls the federation of geographically distributed data centers that function as a single large pool of resources that address any magnitude of workload demand regardless of geography with cost-saving efficiency and assured service performance. Given its high-performance characteristics, the Data Center Without Walls places extraordinary demands on the networks that interconnect geographically separated data centers, especially over the extended overland and undersea network distances associated with global network operators. These demands require end-to-end networks that are highly reliable and resilient, with deterministic bandwidth, low latency, and minimal loss a significant challenge considering the distances involved with global networks. The highly variable characteristics of the 2

3 traffic patterns involved with inter-data center data flows makes this even more challenging. Operators can no longer design undersea networks in isolation if they are to be a part of the cloud computing network and services evolution, since network demarcation hinders cloud adoption. Data Center Without Walls The combination of geographically distributed data centers able to function as an efficient shared resource pool to address any magnitude of workload demand from anywhere across that geography with assured service performance. Global Cloud Backbone Network An integral component of the Data Center Without Walls, leveraging high-capacity, service-oriented architectures between data centers to provide costeffective scale and assured traffic performance. Guaranteed Global Cloud Network Backbone End-to-End Connectivity Required Given this highly unpredictable network environment, the importance of an intelligent global cloud backbone network capable of delivering guaranteed application performance and reliability at least equal to current enterprise data center performance is a prime requirement. Ciena s Global Cloud Backbone Network is an integral component of the Data Center Without Walls concept, whereby Ciena s Performanceon-Demand network capabilities make available a shared pool of switching and transport resources, overland and undersea, that can be efficiently and dynamically allocated to either individual or logically aggregated traffic flows, whenever and wherever required. This highly sophisticated approach to network resource allocation is accomplished through the use of Ciena s V-WAN Hypervisor, which abstracts and virtualizes the Global Cloud Backbone Network resources to allow upper-layer cloud applications and operations to request the allocation of network resources without requiring knowledge of the underlying network. Performance-on-Demand True software-driven networking for the cloud, with automated resource orchestration and provisioning that yields reduced network capacity requirements, predictable service performance, and simpler operations. V-WAN Hypervisor Ciena s network virtualization and automation solution that delivers Performance-on- Demand by automating the allocation of shared network resources among data centers on behalf of cloud operations. The Cloud Opportunity for Global Network Operators Global network operators are approaching cloud-related business opportunities in different ways, with some focusing exclusively on providing reliable connectivity between geographically dispersed data centers spanning terrestrial and submarine networks. These global network specialists provide differentiated network services connecting customers to the cloud over great distances, offering differentiation through lower latency, improved throughput performance, and better network reliability. Through the direct control of their network assets overland and undersea, these global network specialists are in the enviable position of being able to precisely control the end-to-end network SLA. Partnerships among global network operators and cloud computing companies are an increasingly popular approach to capitalize on growing cloud demand the world over. Such partnerships allow global network operators to leverage their cloud computing 3

4 partner s superior expertise quickly in IT hardware and software management, particularly as it pertains to data center design, operation, and service virtualization. Partnering with an experienced IT supplier enables global network operators to enter the market more quickly, gain access to the IT supplier s broad service capability ecosystem, and benefit from their domain-specific brand recognition. Likewise, the IT supplier also benefits from the global network operators expertise and global reach to address new and emerging markets quickly. Vertical Integration is another cloud business model involving the direct ownership and integration of network and data center assets through totally in-house, organic means or via corporate acquisition. Global network operators have long been in the business of providing reliable network connectivity to enterprise customers, and many network operators own data centers that support their network operations. In the last few years, global network operators have sought to leverage their networks and data center facilities to increase overall utilization and enable additional monetization opportunities. A recent Infonetics 2 survey revealed that data center virtualization is the top cloud strategy being pursued by service providers. In fact, 79 percent of respondents replied that they would implement a virtualization strategy by February 2013 a strong endorsement of the anticipated demand for virtualized cloud computing environments for IaaS and other cloud-based solutions, such as PaaS. Enabling Cloud Services Ciena has created an intelligent cloud solution, the Data Center Without Walls, with the goal of significantly reducing net IT resource requirements, enabled by two foundational components: 1. Global Cloud Network Backbone, consisting of the following products: > 6500 Packet-Optical Platform Consolidates Ethernet, SONET/SDH, OTN, and coherent 40G/100G DWDM capabilities into a single platform for the cost-effective delivery of new and existing services over seamlessly connected terrestrial and submarine networks > 5400 Reconfigurable Switching System Intelligent switching platform enabling the practical and cost-effective migration from legacy SONET/SDH-based networks to converged OTN and Ethernet-based networks, with industry-leading, embedded control plane intelligence > Carrier Ethernet Solutions (CES) Full suite of Carrier Ethernet-enabled products that, together, allow the design of robust and cost-effective network designs > OneConnect Intelligent control plane functionality to greatly enhance switching services through the full automation of network operations, improved overall network utilization, differentiated SLAs, and, most importantly from a submarine network perspective, robust network reliability up to six-9s ( %) network availability > OneControl Comprehensive network-wide solution to manage mission-critical networks, including multi-layer intelligence for rapid service turn-up, visibility through protocol layers for proactive/reactive troubleshooting, and the efficient use of network assets. The tight coupling of OneConnect control plane intelligence and OneControl ensures all inherent benefits of Ciena s industry-leading control plane intelligence are made readily accessible. 2. Software-Defined Networking Toolkit, consisting of: > V-WAN Hypervisor Control software that orchestrates access to and performance-assured connections over the global data center network. The V-WAN Hypervisor enables cloud operations to ensure network performance is always aligned with the ever-changing requirements related to virtual machine and data mobility between global data centers, thus providing improved overall economics and application performance. > IT-WAN Orchestration Applications A growing family of optional, standalone software applications for third-party data center systems that monitor their activities, making the appropriate requests to the V-WAN Hypervisor to allocate the correct network resources required to optimize transactions between data centers. These modules are developed by Ciena and/or third parties including EMC VPLEX, VMware vcloud Director, vcenter Management, and Storage DRS. When the Ciena Global Cloud Backbone network is managed under the control of the software-defined networking toolkit, it creates a Performance-on-Demand Global Cloud Backbone network delivering assured network performance at significant capital expenditure savings, compared to over-provisioned IP/MPLS architectures. It is this intelligent network that makes possible the evolution toward a Data Center Without Walls in which IT operations are distance-independent, thereby 2 Data Center Deployment Strategies: Global Service Provider Survey, December 23,

5 providing improved availability, mobility, and collaboration via a network infrastructure characterized by low latency and bandwidth on demand. The advances among service providers in cloud service delivery should be viewed in evolutionary terms. In other words, few global network operators are equipped to jump directly from providing network connectivity to offering high-performance, on-demand cloud computing solutions. However, the ability to connect data centers represents an important first step toward establishing a cloud foundation. Network operators with global reach are in an enviable position, especially if they own the entire network, since they are able to control the end-to-end network SLA precisely and can reach more users. As global network operators start to deploy cloud-based services over their networks that span states, countries, and continents, they must be keenly aware of associated regulatory issues. As such, ensuring compliance with regulatory policies on data remains a key concern to cloud computing. Regulatory policies such as those related to European Union (EU) privacy and Health Insurance Portability and Accountability Act (HIPAA) in the Unites States, for example, place very specific requirements on data availability, integrity, migration, retention, and access, among others. These requirements represent an opportunity for cloud-based network intelligence to implement regulatory-based routing and ensure these stringent policies are respected before data is transferred from one geographic location to another. Assembling the Global Cloud Backbone Network The evolution of the cloud toward expanded enterprise IT utility is driving the creation of a true Data Center Without Walls that comprises multiple global network operator and enterprise data centers, among which various types of traffic can flow. Built with a global cloud backbone network to interconnect these geographically separated data centers with Performance-on-Demand operational capabilities, the ideal network architecture is created to run enterprise-class customer applications effectively. Data centers, regardless of their physical location, operate as a single virtualized entity. Global network operators can (and will) capitalize on the high-growth cloud business opportunity in a variety of ways. Some will focus on providing reliable high-bandwidth data center connectivity over great distances, while others will align with cloud computing specialists to deliver a combined cloud-plus-network connectivity service. The most innovative providers will seek to expand, either organically or through acquisition, their existing ownership of global data center assets to deliver integrated end-to-end IaaS solutions. Regardless of the chosen approach, there will be no shortage of business opportunities, since the worldwide market forecast 3 for public cloud alone is estimated to be $89 billion, with a five-year Compound Annual Growth Rate (CAGR) of 19 percent a significant emerging market indeed. Fully realizing the cloud vision will also entail a series of incremental investments in network and data center infrastructure, IT applications, partnerships, market expertise, and timely strategy execution. Few global network operators currently possess all of the key components required to achieve long-term success in the global cloud services market. The transition to NaaS requires an evolution of today s network technology from static network connectivity to an intelligent and agile infrastructure designed to address the dynamics and economics of evolving cloud operational models. Ciena s global networking toolkit offers the optimum cost at scale and assured service performance for the Data Center without Walls global cloud. By simultaneously minimizing required network capacity while maximizing IT resource utilization, cloud computing demand can be fulfilled anywhere and anytime, overland and undersea, thus breathing new life into tomorrow s global networks. Ciena may from time to time make changes to the products or specifications contained herein without notice. Copyright 2012 Ciena Corporation. All rights reserved. WP Cloud Computing - Economic, Financial & Service Impact on IT Planning Assumptions, December 8, 2011 (Gartner) Networks that change the way you compete.