article page |
1 |
2 |
3 |
4
Although still largely untouched in many commercial applications, this land is not exactly uncharted. Many of today's leading technology developers already exploit hardware and software concepts that support this approach, such as:
-
Parallel processing – A form of computation in which data is broken into smaller pieces so the calculations can be completed simultaneously, in parallel. Parallel processing optimizes the performance gains enabled by multi-core chips. Ericsson has introduced functional language and high parallelism to run some of its most recent switches. Microsoft's Dryad research project offers significant insights into coarse-grained multi-processing technologies that leverage both task and data partitioning in order to move data faster, and with greater stability. For the OSS/BSS, reliance on a single billing engine is eliminated and the concept of a "billing farm" instituted to serve as the workhorse for billing operations.
-
Immutable data – Borrowing from the idea of immutable objects, which are objects whose state cannot be modified after they are created, immutable data includes the concept of writing forward to multiple databases, rather than backwards to a single database. Not only does this streamline data processing, it also contributes additional stability and reliability for complex operations. The OSS/BSS will benefit from parallel programming research in the form of functional languages or functional extensions to existing object-oriented languages, as well as their integration with frameworks that provide underlying support for parallel processing.
|
|
This land is not exactly uncharted. |
|
-
MapReduce – Patented by Google, MapReduce is a software framework that supports distributed computing on large data sets on clusters of computers. MapReduce brings the OSS/BSS significant capabilities in scalability and allows increasingly large data sets to be managed on less costly systems.
-
Cloud computing – Cloud computing or "Platform as a Service" (PaaS) provides the final element of the enhanced OSS/BSS, creating a dynamically scalable architecture that enables new consumption and delivery models for IT services based on the varying needs and business models of the providers served. Clouds can be hosted by a cloud provider (e.g. Microsoft, Amazon), on premises or a combination of the two. Cloud computing provides not just a way of increasing hardware utilization, thus lowering cost, but it also provides a way of commoditizing the software services that support applications.
With the technology table set, telecommunications service providers should expect some remarkable transformations in the performance, commercialization and service offerings related to the OSS/BSS in the coming months and years.
Changing Performance, Delivery and Service
By programming the application to leverage multi-processing, performance gains are not just incremental based on the use of multiple processors and multiple data streams (i.e. four processors and four data streams equals four times as fast). Billing performance under this design model can increase by a factor of 50 with only 5 hardware cores and standard tools. For OSS/BSS providers, this can mean
article page |
1 |
2 |
3 |
4
|
|
|
