A load scheduler for SIP proxy servers: design, implementation and evaluation of a history weighted window approach

Abstract

SUMMARYThe widespread use of Session Initiation Protocol as a signalling protocol has created various challenges. An important one is that its throughput can be severely degraded when an overload happens in the proxy server because of several retransmissions from the user agent. One common approach to overcome this problem is 'load balancing'. A balancer needs to know the status of proxy servers, which are continuously gathered implicitly or explicitly. Implicit methods have averagely less overhead than explicit ones. This paper attempts to prevent throughput reduction by balancing the loads among available proxy servers properly using an implicit mechanism called History Weighted Average Response time. The proposed algorithm is robust because it incurs no extra processing to proxy servers. The novelty of the mechanism is making use of 'response time history' to estimate the load being currently processed on servers. By implementing in a real testbed, throughput and scalability are improved compared with an important state-of-the-art similar algorithm. This improvement stems from no need for modification in SIP protocol, easy implementation and application, simple computations for making decision and no need for extra feedback between servers and load balancer.

OpenSIP: Toward Software-Defined SIP Networking

“…Another approach for dealing with the overload issue is load balancing. This is the distribution of traffic among SIP proxies according to their available capacity by the use of a load balancer [17], [18]. Also, the DNS-based load balancing methods can provide an alternative to in-line load balancing, by associating multiple servers to the same SIP URI [27].…”

“…3) Comparison with Other Approaches: In this subsection, the performance of the OpenSIP Partial is compared with two other wellknown approaches given in [17] and [18]. For this purpose, as shown in Fig.…”

“…A new call is assigned to the SIP proxy with the lowest counter. In [18], to distribute the load with a load balancer, the HWAR and FWAR algorithms are proposed, which use the window and proxy response time to estimate the proxies' load. In the FWAR algorithm, windows sizes are fixed, whereas in HWAR a history of the response times is kept in the windows with the help of a mathematical model.…”

“…[17] and[18] (One stage) Implementation testbeds prepared to compare OpenSIP Partial with[17] and[18] SIP proxies' peak throughput in various methods with eight SIP proxies SIP proxies' response time in various methods with eight SIP proxies…”

Load-Balanced and QoS-Aware Software-Defined Internet of Things

“…Table 4 illustrates the details of the network configuration used to perform the experiments. The Load Balancing by Server Response Time (LBBSRT) [50], History Window Weighted Average Response Time (HWAR) [51], and Transaction Least-Work-Left (TLWL) [52] methods are the most important comparison algorithms in this regard. LBBSRT uses the controller to obtain the response time of each server in order to select a server with minimum the response time.…”