USENIX Association Proceedings of the 5 th Annual Linux
نویسندگان
چکیده
The current 2.4 Linux network stack is based on a solid multi−processor−threaded implementation, known as softnet, which has been validated over the last 2 years. Linux is also used extensively in Internet packet forwarding services such as firewalls and routers etc. Linux has a very modular packet processing framework based on the (ingress and egress) Traffic Control Framework as well as the Netfilter IP firewalling architecture. While the Linux framework for "plugging in" packet services is very flexible, and very much proven in the real world, for Linux to be considered a strong Network Operating System, robustness under all conditions (including severe overload) is a key requirement that must be met. Our work is to further improve Linux to have the following attributes: Robustness at any input rate and any number of input devices. Controlled and low Latency. cure packet reordering that is inherent with SMP support. Provide fairness in greedy network when supporting many interfaces under overload We discuss the problems, solutions and provide experimental results in our attempts to deal with these issues. While the focus is on using a PC as a router, the solutions provided are applicable to Linux in use in all aspects as a network device (such as a server).
منابع مشابه
USENIX Association Proceedings of the 4 th Annual Linux
Rights to individual papers remain with the author or the author's employer. Permission is granted for noncommercial reproduction of the work for educational or research purposes. This copyright notice must be included in the reproduced paper. USENIX acknowledges all trademarks herein.
متن کاملUSENIX Association Proceedings of the 4 th Annual Linux Showcase
This paper presents a multi-threaded BLAS library for dual SMP Intel computer running Linux. We present simple techniques to obtain parallelism for BLAS call transparently from the client program. We discuss some synchronization methods available under Linux, show performances results for a representative set of BLAS and for a high level linear algebra kernel. We then explain some key points on...
متن کاملUSENIX Association Proceedings of the 4 th Annual Linux Showcase
We evaluate and compare the performance of LAM, MPICH, and MVICH on a Linux cluster connected by a Gigabit Ethernet network. Performance statistics are collected using NetPIPE which show the behavior of LAM/MPI and MPICH over a gigabit network. Since LAM and MPICH use the TCP/IP socket interface for communicating messages, it is critical to have high TCP/IP performance. Despite many efforts to ...
متن کاملUSENIX Association Proceedings of the FREENIX Track : 2002 USENIX Annual Technical
We introduce a simple sequential write benchmark and use it to improve Linux NFS client write performance. We reduce the latency of the write() system call, improve SMP write performance, and reduce kernel CPU processing during sequential writes. Cached write throughput to NFS files improves by more than a factor of three.
متن کاملUSENIX Association Proceedings of the 4 th Annual
Rights to individual papers remain with the author or the author's employer. Permission is granted for noncommercial reproduction of the work for educational or research purposes. This copyright notice must be included in the reproduced paper. USENIX acknowledges all trademarks herein.
متن کاملUSENIX Association Proceedings of the 4 th Annual
Rights to individual papers remain with the author or the author's employer. Permission is granted for noncommercial reproduction of the work for educational or research purposes. This copyright notice must be included in the reproduced paper. USENIX acknowledges all trademarks herein.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2001