On the performance of IP-forwarding for multicore multiprocessor Linux hosts

Salah, Khaled; Hamawi, Mohamed; Hassan, Yasser

doi:10.1049/iet-com.2010.0057

Cited by 7 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The research that is related to this paper falls under 2 main categories: (1) the performance prediction of packet forwarding on software routers and (2) cache contention in multicore processing systems. In the first category, a number of proposals have analytically evaluated the performance of software routers on the basis of Linux network stack in different ways and come up with what factors influence the performance bottleneck. In the work of Rojas‐Cessa et al, the processing overhead of an operating system encapsulated in a virtual machine has been considered as the throughput penalty of a virtual software router (referred to a software router running in a virtual environment), whereas a software router can fully use the hardware resources of a server.…”

Section: Related Workmentioning

confidence: 99%

“…In addition, the performance of a virtual software router can also be improved with the adoption of multicore processing system. The performance of packet‐forwarding function, in fact, depends on the host's resources and its configuration . The study of Salah et al identifies hardware and software architectural bottlenecks that can significantly impact the performance of packet forwarding.…”

Section: Related Workmentioning

confidence: 99%

“…The performance of packet‐forwarding function, in fact, depends on the host's resources and its configuration . The study of Salah et al identifies hardware and software architectural bottlenecks that can significantly impact the performance of packet forwarding. On the other hand, recent studies of software routers have unanimously affirmed that the CPU speed and Ethernet bandwidth are key factors influencing the performance bottlenecks.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Towards performance prediction of multicore software routers

et al. 2017

View full text Add to dashboard Cite

Rapid advances in network function virtualization technologies have led to the emergence of flexible and scalable embedded data plane processing functions, eg, packet forwarding, on commodity hardware. However, performance prediction of softwarized network functions on such shared resources is challenging but very important for obtaining the full benefits of network function virtualization. This paper addresses the problem of performance prediction for multicore software routers and reveals a key technique to achieve high accuracy. Motivated by observations, we first analyze how many CPU cycles are spent for forwarding a packet in multicore processing systems. Our prediction model of the CPU usage based on cache contention can capture its nonlinear dilation scaled by the number of CPU cores-called dilated CPU consumption (DCC). We validate the accuracy of the DCC model with measured data. On the basis of the DCC model, we develop 2 performance prediction algorithms to predict the maximum throughput of the packet-forwarding function corresponding to the assigned resources. The first algorithm includes CPU utilization statistics (called DCC-u), while its simplified version (called sDCC) does not require CPU utilization statistics. We validate the proposed models through the exhaustive set of experiments under various amounts of assigned resources (ie, CPU speeds and the number of CPU cores) and different traffic loads (ie, a large and small number of traffic flows). The results show that DCC-u and sDCC achieve high accuracy under various amounts of resources and traffic conditions. Remarkably, they improve the precision of estimation from that of the existing techniques by up to 105% to 163%. KEYWORDS functional areas, multicore software router, network function virtualization, performance management, performance prediction 1 Int J Network Mgmt. 2017;27:e1966.wileyonlinelibrary.com/journal/nem

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Towards performance prediction of multicore software routers

et al. 2017

View full text Add to dashboard Cite

show abstract

“…High-bandwidth environment, however, can receive millions of frames per second which will make the overall performance of the system to suffer. In order to adapt to a high-bandwidth environment, Linux Kernel 2.6 introduced a new interface called NAPI [11,12]. NAPI is one of the ways to mitigate interruption caused by such a burst arrival of frames.…”

Section: Design Of Priority Processing Device Drivermentioning

confidence: 99%

Parallel implementation of real-time communication and IP communication by using multiple ring buffers

Ueda

Kikutani

Yakoh

2014

2014 10th IEEE Workshop on Factory Communication Systems (WFCS 2014)

View full text Add to dashboard Cite

Recent Linux [7] kernel introduced an interrupt mitigation method called NAPI to reduce the overhead of interrupt handling procedures caused by frame arrival and transmission. However, NAPI caused serious delay and jitter in such procedures. On the other hand, many conventional real-time communication mechanisms were implemented independently to the existing network stack, and most of such mechanisms disabled the stack. Therefore, it is difficult to utilize both of real-time communication and IP communication in parallel.This paper proposed a parallel implementation method to realize both of real-time communication and IP communication simultaneously in a single computer unit. This method implement real-time network stack and IP network stack in parallel by using multiple queue functionality in a NIC (network interface card). Also, this method applied the interrupt mitigation method only to the IP stack. Therefore, this method achieved to reduce the overhead of the wide bandwidth of IP traffic, also to keep delay and jitter of real-time traffic enough small. Some experiments were conducted to measure the performance of the proposed method. The RTT (round trip time) delay of realtime communication was kept lower than 100 µs even in high traffic load of 800 Mbit/s with small jitter, while the RTT of IP communication varied between 100 µs to 20 ms. This paper introduced the design, implementation, and evaluation of the proposed method.

show abstract

“…To keep continuous connectivity, a user terminal has to handoff its connection from one AP to the next when the user moves in the area [6] .…”

Section: Wifi Wireless Communication Modulementioning

confidence: 99%

A novel remote and virtual driving system based on WiFi communication

Cao

Huang

et al. 2014

2014 4th IEEE International Conference on Information Science and Technology

View full text Add to dashboard Cite

Unmanned intelligent vehicles become a hot spot of the world vehicle engineering and growing innovatory motivation in the automotive industry. By the Wifi communication, a novel remote and virtual driving system is developed in this paper. AM3359 processor equipped with the Linux is used in the remote and virtual driving system. Real-time environmental information of the car is captured by in-vehicle camera. The car's environmental information is transmitted to the client by the WiFi network. The steering wheel and other equipments are used to realize the remote real-time control of the car. The remote and virtual driving system consists of data collection and transmission module, system control module, system security module, power management module, remote communication and monitoring module, wireless communication module, and so on. PC terminal is established by a video display terminal, steering wheel, and other control equipments. The remote real-time operating experiments are performed. The experimental results show that users can monitor the operational status of the vehicle and the remote real-time simulation driving is achieved. The results verify that the developed remote and virtual driving system is effective, reliable, and real-time. Index Terms -Linux; WiFi; video transmission; remote and virtual driving system

show abstract

On the performance of IP-forwarding for multicore multiprocessor Linux hosts

Cited by 7 publications

References 7 publications

Towards performance prediction of multicore software routers

Towards performance prediction of multicore software routers

Parallel implementation of real-time communication and IP communication by using multiple ring buffers

A novel remote and virtual driving system based on WiFi communication

Contact Info

Product

Resources

About