At MWC this year I saw a budding constellation of similar thinking around “micro-service” innovation. This is a very positive trend.
Every service provider wants greater agility, easier innovation, lower innovation costs and reduced ongoing maintenance burdens. No surprise here. Yet in our industry, most innovation methods and supporting software are essentially one-offs, with significant development times & point to point integration. This leads directly to slow rollout, high costs, and the “gift that keeps on giving” – the need to continually maintain and re-integrate these complex processes. Its one reason that such a high proportion of IT budgets are expended on simply keeping the existing environment running – leaving far less money and manpower for innovation.
Appledore Research is not alone in promoting the need to switch to more efficient methods: typically those of service-oriented (“micro-services” to use the latest buzzword) components, composed into complete services, and fostering a high degree of re-use. By re-using “services”, this allows orders-of-magnitude improvements in agility and cost (its already built). Moreover, if you re-use a service object, you maintain it and integrate it ONCE. So that ongoing expense comes down steadily over time. This gift continues giving in a much more positive way.
Enough theory. At MWC I saw two very different, complementary and interesting approaches that begin to show the way forward. One, Ribbon’s “Kandy”, is reasonably well established. The other, Amdocs’ “microservices360” is just announced on day one of MWC and is therefore less well documented.
Functionally, the two could hardly be more different. Kandy is in the network communications space while microservices360 is in the BSS/OSS space. Kandy is a set of hosted communications services that can be “mashed up” and connected to one or more CSPs and enterprises environment. It can be used to build entirely new CSP services, or to integrate together existing service in two or more providers (or enterprises). Amdocs’ microservices360, on the other hand, is a services-oriented BSS/OSS development framework intended to deliver better cost and agility to new OSS/BSS innovation. To quote Amdocs, it is “a deployment and development platform based on vetted and tested open source tools”, and is delivered as part of new Amdocs functionality. Once in place, either Amdocs or the CSP can add new micro-services and use the microservices360 environment to integrate disparate systems or create a new functional flow (e.g.:provision a new service, using resources across existing OSSes as well as new micro-service functionality).
In concept however, the two could not be more similar: each is a library of re-usable, loosely coupled functionality, an API environment and each brings an agile, service oriented approach to innovation. Moreover, each is well suited to integrating existing, disparate environments: Kandy can link into many CSP and enterprises environments acting as a sort of gateway when needed, and microservices360 serves a similar function between existing, typically older tech, OSSes and BSSes.
Some day we will have entirely modern, service-based OSS, BSS and network (e.g.: orchestrated services in the network). We will have well structured catalogs of customer- and network- facing functionality that can be strung together by NFV-Os and service orchestrators. But today we must do our best to make the existing, hybrid infrastructure agile. We have a large embedded base of OSS and network functionality, small pockets of new functionality in both domains, and the need to innovate and link existing functionality more efficiently and to build, step-by-step, toward an agile future. Services-Oriented frameworks, along with essential “services” are the best path forward.
Appledore Research believes that these two offers highlight a promising trend and encourage Service Providers and software suppliers to continue the momentum – both in the network and in the OSS/BSS/MANO+ environments.