Developing Highly Scalable Solution for SON
t the Client
The client is a start-up in the field of Wi-Fi (Wireless Networks) solving the problems of self-optimizing networks for various product vendors
The customer wanted ALTEN Calsoft Labs to develop a centralized server hosted on cloud (public/private cloud platforms) which can do the following:
- Scale infinitely for serving millions of access points at the same time
- Provide a scalable storage to host the statistics generated by all the Wi-Fi Aps
ALTEN Calsoft Labs re-architected the SON server (PULL model was changed to PUSH model) thus making it stateless so that the services can be easily scalable with the integration of MOM service running between the client access points. The SON server offered high availability, scalable infrastructure and acted as a load balancer to balance the load across multiple servers.
- Making the server stateless, it is possible that any server can process the requests originated by any AP and can scale infinitely along with high availability offered by the MOM service sitting between server and the Aps
- The usage of RabbitMQ, Celery and MongoDB provided a scalable and robust solution.
- Platform: Hadoop
- File System: Hadoop Distributed File System
- Paradigm: MapReduce
- Machine Learning Tools: Mahout
- Language: R
Let’s fast-track your next big idea