ISE03-3: Towards Stateless Core: Improving SIP Proxy Scalability

Cortés, Mauricio; Esteban, Jairo; Jun, Hyewon

doi:10.1109/glocom.2006.193

Cited by 12 publications

(10 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Balasubramaniyan et al [5] propose a dynamic distribution algorithm to continue to maintain state and yet improve throughput in a proxy farm. Cortes et al [9] propose that SIP proxies in an IMS infrastructure should be configured with stateless and stateful functionality, allowing them to adapt to network and CPU conditions dynamically. Singh et al [25,26,27] present a reliable, scalable and interoperable Internet telephony architecture for user registration, call routing, conferencing and unified messaging using commodity hardware.…”

Section: Related Workmentioning

confidence: 99%

Improving authentication performance of distributed SIP proxies

Dacosta

Balasubramaniyan

Ahamad

et al. 2009

Proceedings of the 3rd International Conference on Principles, Systems and Applications of IP Telecommunications

View full text Add to dashboard Cite

The performance of SIP proxies is critical for the robust operation of many applications. However, the use of even light-weight authentication schemes can significantly degrade throughput in these systems. In particular, systems in which multiple proxies share a remote authentication database can experience reduced performance due to latency. In this paper, we investigate how the application of parallel execution and batching can be used to maximize throughput while carefully balancing demands for bandwidth and call failure rates. Through the use of a modified version of OpenSER, a high-performance SIP proxy, we demonstrate that the traditional recommendation of simply launching a large number of parallel processes not only incurs substantial overhead and increases dropped calls, but can actually decrease call throughput. An alternative technique that we implement, request batching, fails to achieve similarly high proxy throughput. Through a carefully selected mix of batching and parallelization, we reduce the bandwidth required to maximize authenticated signaling throughput by the proxy by more than 75%. This mix also keeps the call loss rates below 1% at peak performance. Through this, we significantly reduce the cost and increase the throughput of authentication for large-scale networks supporting SIP applications.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving authentication performance of distributed SIP proxies

Dacosta

Balasubramaniyan

Ahamad

et al. 2009

Proceedings of the 3rd International Conference on Principles, Systems and Applications of IP Telecommunications

View full text Add to dashboard Cite

show abstract

“…Is has also been proposed to configure a proxy such that state information creation is limited to a fraction of messages (such as INVITE and BYE) based on message type, CPU load and link quality parameters and stateless proxies be used in case of having low packet loss probability or when a reliable transport protocol such as TCP is used [12].…”

Section: B Common Deployment Scenariosmentioning

confidence: 99%

“…Proxy servers which are used for forking calls, NAT traversal, retransmission handling, and accounting, must be configured stateful and are typically located at the edge of network. Instead, the translator and forwarding proxies, located in the core of the network must be configured stateless [12].…”

Section: B Common Deployment Scenariosmentioning

confidence: 99%

Configuration of a SIP Signaling Network: An Experimental Analysis

Homayouni

Azhari

Jahanbakhsh

et al. 2009

2009 Fifth International Joint Conference on INC, IMS and IDC

View full text Add to dashboard Cite

Session Initiation Protocol (SIP) [1] is adopted as the main signaling protocol for the IP Multimedia Subsystem (IMS). Large deployment of SIP servers in Next Generation Networks (NGN) requires studying the behavior and performance of SIP servers under both normal and overload conditions, which may happen in such networks with a vast number of users. In this paper a measurement-based study is performed to assess SIP proxy server performance under different configurations, and for different underlying transport protocols, for both inter-domain and intra-domain calls. Test results show the effectiveness of configuration and distance on SIP signaling performance. Therefore a great attention must be paid during SIP servers deployment.

show abstract

“…Current discussion in the IETF [9][16] centers on creating an overload control framework and adopting appropriate new headers to convey additional information beyond a simple 503 response. The approach taken in [5] leverages the fact that throughput of a SIP proxy is higher when processing requests in a transaction-stateless manner, and thus their solution consists of handling a subset of requests statelessly during the onset of congestion, thus trying to avoid overload. In contrast to the aforementioned work, our focus is on creating a mechanism in the form of a programmable engine that enables user-defined policies to be executed efficiently, without modifications to the SIP server.…”

Section: Related Workmentioning

confidence: 99%

A programmable message classification engine for session initiation protocol (SIP)

Acharya

Wright

2007

Proceedings of the 3rd ACM/IEEE Symposium on Architecture for Networking and Communications Systems

View full text Add to dashboard Cite

Session Initiation Protocol (SIP) has begun to be widely deployed for multiple services such as VoIP, Instant Messaging and Presence. Each of these services uses different SIP messages, and depending on the value of a service, e.g. revenue, the associated messages may need to be prioritized accordingly. Even within the same service, different messages may be assigned different priorities.In this paper, we present the design and implementation of a programmable classification engine for SIP messages in the Linux kernel. This design uses a novel algorithm that in addition to classifying messages can extract and maintain state information across multiple messages. We apply the classifier for overload control using operator-specified rules for categorizing messages and associated actions, augmented with a protocol-level understanding of SIP message structure. When faced with loads beyond their capacity (e.g., during catastrophic situations and major network outages), SIP servers must drop messages. It is therefore desirable that the server process highvalue messages in preference to lower-value messages. We evaluated our in-kernel classifier implementation with an open source SIP server (SER) for such an overload scenario. The workload consists of a mix of call setup and call handoff messages, and the classifier is programmed with rules that prioritize handoffs over call setups. We show that, while SER can process about 40K messages/sec (in a FIFO manner), our classifier can examine and prioritize 105K messages/sec during overload. With the classifier operating at peak throughput, SER's processing rate drops to 31.6K messages/sec, but all of the available high-value messages are processed.

show abstract

ISE03-3: Towards Stateless Core: Improving SIP Proxy Scalability

Cited by 12 publications

References 4 publications

Improving authentication performance of distributed SIP proxies

Improving authentication performance of distributed SIP proxies

Configuration of a SIP Signaling Network: An Experimental Analysis

A programmable message classification engine for session initiation protocol (SIP)

Contact Info

Product

Resources

About