Sponsored By
The virtualization of networks and the transition to cloud structures--in which bandwidth, network capabilities, services and storage assets are managed from high level repositories and deployed on an as-needed basis--is changing management structures as well.
Operations support services/business support services (OSS/BSS) are being called on to perform a litany of tasks unheard of when networks were static and highly regimented. Now, a carrier’s OSS/BSS must see the entire end-to-end path of a transmission, no matter what route it takes, and provision it on a real time basis.
It is nothing short of a reinvention. AT&T's approach is to re-architect the OSS/BSS into containers built on a trio of foundations: platform/application, distribution and infrastructure.
Reworking the OSS/BSS infrastructure on the fly is no easy task. Telco Transformation spoke to Douglas Nassaur and Sorabh Saxena about the AT&T’s OSS/BSS journey. Nassaur is the telco’s general manager and lead principal technical architect in the technology design and architecture organization. Saxena is AT&T’s senior vice president of software development and engineering for technology development. Their answers have been combined in this Q&A.
AT&T: We anticipated the need several years ago to establish a more flexible and efficient computing platform to align with our diverse portfolio of products and services. Leveraging a platform approach supported by API enabled services, we established the foundation for providing an integrated service experience for our customers.
The platforms and standards we established are pivotal to the transformation of our OSS/BSS. They afford us the ability to incorporate and adopt new innovation as it occurs across cloud, container and micro-service componentry to extend our platforms and frameworks.
We are transforming OSS/BSS in a two-step process. The first is the decomposition of the software that implements business and technical functionality into container-based deployment models and, secondly, the refactoring of individual services into true microservice architectures.
It is paramount to our success that we view the container/cloud-based value chain through its three key foundations; infrastructure, distribution and platform/application. Our ultimate goal is to “right-size” the resource spend associated with specific business or technical functionality. The foundations contribute the following:
Much like the delineation between electricity generation, power distribution, transformation and consumption in the power company model, each of the three major foundations of containers contributes to the overall benefits of cloud and containers.
OSS/BSS is evolving from a suite of independent applications, each with interfaces between them, to a set of microservices supported by a platform of common services, frameworks and runtimes. Each key function of the application is decomposed into a functional element that publishes itself to the platform. Any other element that wishes to interact simply asks the platform to provide the service--significantly reducing time, cost and complexity.
In much the same way that countries and cultures with different languages and customs find ways to interact, platform-based services will interface across global boundaries. Industry-standards will define vendor, technology and stack neutral interface agreements and will facilitate interoperability while avoiding vendor and technology lock-in.
TT: Is your existing OSS hampering your ability to offer on demand services? If yes, why?
AT&T: Our focus is on providing an integrated suite of services across multiple service domains. As a result, traditional OSS/BSS demands levels of integration and interoperability across applications that traditionally did not interact. Our approach has been to invest in unified platform services, frameworks and runtimes to facilitate interoperability instead of pursuing individual application point-to-point integrations.
TT: Which of the traditional OSS and BSS categories, such as service assurance, revenue management, and customer experience management, is your company more likely to be investing in over the next two years?
AT&T: Our focus is end-to-end transformation and digitization. Hence, all will benefit from our unified platform, common frameworks and runtimes approach. Platform services will benefit from container-based deployment at scale and functional level elasticity to “right-size” expense to point-in-time demand.
TT: In an increasingly competitive landscape, can improved OSS/BSS be a big differentiator for a service provider?
AT&T: The ability to facilitate interoperability across traditional OSS/BSS functions in a timely and cost-effective manner will be the true differentiator. In much the manner AT&T was viewed as the voice/data switchboard that connected people and machines, we will be seen as the digital cross-connect of content and capability across a producer, provider, and customer in a complex multi-cloud landscape. AT&T will deliver integrated and unified experiences to the customer in real time, wherever they are and whatever they would like to do.
TT: Is there a need for more convergence further down the OSS stack?
AT&T: The true need is a standard set of interface agreements at each foundation level. Freedom of choice in integration and implementation of these interface agreements is vital as is the need for them to remain technology, vendor and stack agnostic. This approach enables freedom to adopt open source innovation and to replace modular elements when and where appropriate.
TT: What are the challenges for integrating OSS across the different network domains such as CDMA, LTE, 5G, IoT, etc.?
AT&T: Recognition of the three foundations introduced earlier (infrastructure, distribution and platform/application) is vital. If truly implemented, these three would allow cross services to be more easily integrated and interoperable. Additionally, we believe transformation of traditional security and operational paradigms are also needed. For example, traditional security models employ exclusionary--inside firewall good, outside firewall bad-- security models. The security model going forward must exhibit inclusionary: We know who or what you are, the role you play, where you happen to be and what you are allowed to do in that context of security models. These types of pivots support the need and direction toward unified platforms, frameworks and runtimes and industry standard interface agreements that are stack, technology and vendor agnostic.
AT&T: The framework of the future is multidimensional. It must address the foundations necessary to better define, distribute and execute software across complex environments. It must also contemplate the need for separation of concerns and span of control where diverse networks and multiple providers and technologies are in play.
Standards must express and enforce behavior while being flexible enough to be implemented using different tools, and even accommodate physical and software-defined resources and functions. Simply put, standards and frameworks must assure a consistent, predictable and repeatable customer experience across diverse infrastructures, networks and providers. Hence, the standard should be very prescriptive when describing interface agreements which govern behavior while being flexible in how the interfaces are implemented technically.
In addition, the unified service landscape will transcend traditional barriers (5G, Cloud, IoT, LTE, big data, etc.) and require an OSS/BSS perspective that supports a “cascading call path” – the ability to enable and manage a transaction across all end-to-end enablers with an unprecedented level of visibility and control. This will require separation of concerns across infrastructure, distribution networks and platforms/applications across public/private enablers with clear interface agreements and governance to respect preferences, sensitivities and tolerances required to meet SLAs and security/regulatory concerns/requirements.
TT: What challenges does the dynamic nature of virtualized networks bring to traditional service assurance and performance monitoring?
AT&T: It shifts the focus from independent tools layered at the device/appliance/network facility to a focus on the services being provided and the call path/transaction path between the producer/provider/customer (infrastructure “upview” to functional “downview”). We are fundamentally shifting the focus of tools from managing the heartbeat, blood pressure and respiration of a black box device to instrumenting how well the overall business or technical function performed across all its resource enablers.
We historically assume all vital organs have been located in the same body. The go-forward model may place the lungs in a location and environment completely different from the heart and circulatory system while making them still appear to be physically connected.
Related posts:
— Carl Weinschenk, Contributing Writer
Telco Transformation
