A number of years back, my product team engaged with a Tier 1 solution provider. They wanted to use our IMG media gateway as part of their solution, but with a condition. They had limited rack space, so they wanted to use an existing server to manage our device. Up until then, we required customers to load our element management system (EMS) software onto a dedicated Linux server. Instead, our customer asked us to take our EMS software and package it to run on a virtual machine. Our team investigated and were able to port both the underlying Linux layer and the EMS application for use on a Xen virtual machine. Voila! Our software was now virtualized and our customer was able to re-use their existing server to manage the IMG gateway.
That was my introduction to virtualization, but this approach quickly became much more important. Just a few months later, other customers asked us to port our EMS software to work within the VMWare virtual machine environment. There were immediate benefits. The EMS running directly on server hardware required qualification of new servers roughly every two years, an arduous process which became more difficult over time. By contrast, the virtual EMS (which we shortened to call the VEC), would run on a VMWare virtual machine and we were isolated from any server changes the customer might make. The VEC was also a software based product, so we offered it for much less than $1000 retail price vs. the $3000+ price point of a server based version. Over the next several years, more and more customers moved to the virtualized version of software and the demand for the server version declined.
A couple of years ago, I was asked to take over a new software-based load balancer (LB) product developed by a Dialogic software team in the United Kingdom. The back story here had some similarities to my earlier experience. The team was working with a major customer who really liked their software-based media resource broker (MRB), but had issues with the LB product offered by a major market player. The team built the software load balancer so that it could run either directly on a server or on a virtual machine. When we launched the product for use by all customers, our Sales Engineering team loaded the software onto their laptops using a commonly available virtual software program and were immediately able to set up prototype sessions and adjust configurations via the software’s graphical user interface. So the LB software was virtualized from the beginning. This was part of an overall trend within Dialogic, as more and more of the software-based components of products were converted for use in virtual environments.
In the early days, virtualization in telecom was mainly for software tools like user interfaces and configuration, but that is now changing in a major way. The LB product from Dialogic runs in a totally virtual mode, so that operations as diverse as configuration and balancing streams of protocols as diverse as HTTP and SIP all are supported, along with very robust security. In the telecom industry, virtualization is being used in several different ways as part of a sea change where the new approach to scalability involves building additional capability and resiliency by adding new instances of software. In turn, this drives the need for new types of orchestration software, which can manage operations in a world where the new paradigm requires creating, managing and deleting instances based on real time needs.
In my next post, I’ll talk about other ways that virtualization is being used as a key principle for building out telecom operations in a variety of Cloud environments. Virtualization is still a relatively young technological movement, but it has already helped spawn some surprising developments.
If you participated in the evolution described here, please feel free to weigh in with your comments. If you’d like to explore strategies on how to evolve your application solutions or other communications products in this rapidly changing business environment, you can reach me on LinkedIn.