Okay, time to discuss one of the most oft claimed, but rarely proven, item in the inter-carrier voice OSS & BSS solution space: Scalability. Scalability is critical, but when, why and how are not often explained. Here at GCS, we have spent thousands of man-hours figuring out how we provide scalability. Why so many hours? Because it is hard to do.
So, when do you need scalability?
According to Wikipedia, Scalability is defined as “the capability of a system, network, or process to handle a growing amount of work.” That’s a pretty good definition, but we have refined the definition as follows:
Scalability: The ability of a system or process to handle a growing amount of work without materially impacting the time it takes to perform the work.
In our opinion, “ without materially impacting the time it takes to perform the work”, is a critical component of any truly scalable system or process. It’s easy to add hardware and therefore expand the capacity of a system or process to perform a particular task or set of work. But, to do this in a way that doesn’t materially impact the time to perform that task or set of work, is what is compelling. What we are talking about here is speed. From our perspective, the speed at which a system processes a greater volume of work is just as important as the volume it can process.
Typically, Computers work on a traditional FIFO model (First in, First Out) approach and they complete tasks in that order. With today’s modern technology, we can create parallel processes (multi-threads) that helps create simultaneity in processes and that does help improve speed of performance. But here, again, scalability is provided by hardware. Improving the design of the software is what needs to be done in addition to the hardware.
Since the inception of the storing of data, companies have employed data scientists and architects who have researched and explored how to improve the speed at which a system operates. Over the past 5 years this has resulted in what we now call “Big Data”. Big Data encapsulates everything about processing massive amounts of information in a data warehouse that can be accessed, processed and utilized by an application.
GCS has been leveraging a Big Data solution based on Oracle and Infobright that allows us to process billions of records quickly, yielding enormous insight and performance improvements to all aspects of inter-carrier voice. We have helped customers reduce their CDR processing (reporting, analysis, invoicing, etc.) from weeks to days to hours. Some of our largest customers are generating billions of CDRs per month and they are able to analyze and react to business conditions almost immediately because of the speed of our solution. We designed our software to maximize the hardware resources available so we can handle volume and maintain speed.
So, the next time you think about Scalability make sure you take a few seconds and think about the speed.