Improving authentication performance of distributed SIP proxies
Proceedings of the 3rd International Conference on Principles, Systems and Applications of IP Telecommunications
The impact of TLS on SIP server performance
Principles, Systems and Applications of IP Telecommunications
Auto-scaling emergency call centres using cloud resources to handle disasters
Proceedings of the Nineteenth International Workshop on Quality of Service
Network convergence and QoS for future multimedia services in the VISION project
Computer Networks: The International Journal of Computer and Telecommunications Networking
The impact of TLS on SIP server performance: measurement and modeling
IEEE/ACM Transactions on Networking (TON)
| Hi-index | 0.00 |
A growing class of applications, including VoIP, IM and Presence, are enabled by the Session Initiation Protocol (SIP). Requests in SIP typically traverse through multiple proxies. The availability of multiple proxies offers the flexibility to distribute proxy functionality across several nodes. In particular, after experimentally demonstrating that the resource consumption of maintaining state is significant, we define the problem of state distribution across multiple nodes when the goal is to increase overall call throughput. We first formulate this as an optimization problem and then derive a distributed algorithm from it. This distributed algorithm leads to the design and evaluation of SERvartuka, a more scalable SIP server that dynamically determines the number of SIP requests for which the server is stateful while delegating state maintenance for the remainder of the requests to a server further downstream. This design is in contrast to existing SIP servers that are statically configured to either be stateless or stateful and therefore result in sub-optimal call throughput. We implement SERvartuka on top of OpenSER, a commercial SIP proxy server and measure performance benefits of different server configurations. An example of our results is a 20% percent increase in call throughput when using our algorithm for a configuration of two servers in series.