> On 06 Dec 2014, at 17:57, Igor Romanovich <[email protected]> > wrote: > > Thanks a lot Alfredo for your great support! One question remains: Are > PF RING ZC queues directly corresponds to the physical queues of network > interfaces?
Yes > If not, are the physical queues utilized? > > On 12/04/2014 05:26 PM, Alfredo Cardigliano wrote: >> >>> On 04 Dec 2014, at 14:51, Igor Romanovich <[email protected]> >>> wrote: >>> >>> You mean that I don't need to create the RSS queue's and they are >>> created when the driver is loaded? >> >> You should use: >> insmod ./ixgbe.ko MQ=1,1,1,1 RSS=8,8,8,8 >> >>> Ok, so what name device_name should I use when opening the device? For >>> example to open queue K on ethX, should I use "zc:ethX@K"? Right? >> >> Correct. >> >> Alfredo >> >>> >>> On 12/04/2014 05:08 PM, Alfredo Cardigliano wrote: >>>> Igor >>>> the number of RSS queues must be specified at insmod time, please take a >>>> look at load_driver.sh >>>> >>>> Alfredo >>>> >>>>> On 04 Dec 2014, at 13:18, Igor Romanovich <[email protected]> >>>>> wrote: >>>>> >>>>> Hi, >>>>> >>>>> Thanks for your response Alfredo. I am planning to use PF_RING ZC. I >>>>> know how to load the network driver with RSS enabled, but I'm not sure >>>>> on how to create a new RSS queue. >>>>> Is it listed here? http://www.ntop.org/pfring_api/pfring__zc_8h.html >>>>> Which function is it? >>>>> >>>>> On 12/04/2014 02:55 PM, Igor Romanovich wrote: >>>>>> Hi, >>>>>> >>>>>> PF_RING is great software, but I'm a little puzzled on how to use it. >>>>>> Suppose an application which has 6-8 CPU-intensive packet producer >>>>>> threads. Now we want to write the packets produced by these threads to >>>>>> some multiple 10G network interfaces... >>>>>> >>>>>> What is the ideal architecture using PF_RING for high-throughput >>>>>> transmission? >>>>>> >>>>>> These are some options: >>>>>> >>>>>> 1- There is a consumer thread for each interface that reads the packets >>>>>> from a queue, and writes packets to the interface. Producer threads >>>>>> write packets to the queues of the consumer threads. >>>>>> >>>>>> 2- Each producer writes it packets to some ring of each network >>>>>> interface (without any consumer thread). I know Netmap supports that. >>>>>> PF_RING has a similar concept of rings, but I'm not sure that it really >>>>>> utilizes that network device rings. >>>>>> >>>>>> 3- Using a pf_ring cluster and multiple fan-in queues, and leaving the >>>>>> fan-out of the packets to pf_ring. This option is not ideal for me as >>>>>> pf_ring apparently doesn't support per flow fan-out. >>>>>> >>>>>> What architecture do you guys recommend? >>>>>> Please note the following in your answer: >>>>>> - What's the architecture? >>>>>> - Which of PF_RING ZC or PF_RING DNA you are recommending >>>>>> - Are you using non-blocking I/O or not? >>>>>> >>>>>> Thanks for your support. >>>>>> >>>>>> Best Regards >>>>>> >>>>> _______________________________________________ >>>>> Ntop-misc mailing list >>>>> [email protected] >>>>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc >>>> >>>> _______________________________________________ >>>> Ntop-misc mailing list >>>> [email protected] >>>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc >>>> >> _______________________________________________ Ntop-misc mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop-misc
