[srslte-users] Throughput tests

Kewin Rausch krausch at fbk.eu
Wed Feb 14 10:28:28 UTC 2018


Hey Dimitri,

having lot of losses is a way to signal that the medium you are using is
not behaving well; as Ismael said, check carefully the level of the signal.
Interference can create a real mess there. If you have USRP try to put some
antennas. I use it lot of time without them, to avoid catching/interfere
with other experiments here in the lab, but this has countereffect on
performances.

HARQ processes of course will retry to carry on their jobs, and probably
some of the packets you are receiving correctly have been re-transmitted
(due to low signal/interference).
Using the BW/subframes to do all those re-tx will eat up the throughput.

You need to be sure that the signal between eNB and UE is not disturbed and
optimal to carry the experiment.
Once you got that, then it means that something else is wrong... and you
can start from there to analyze EPC and eNB.

​Kewin​


On Wed, Feb 14, 2018 at 11:07 AM, Ismael Gomez <
ismael.gomez at softwareradiosystems.com> wrote:

> What do you see in the UE and ENB metrics, how are the dl and ul snr
> levels?
>
>
> El mié., 14 feb. 2018 11:01, <dmakris at iit.demokritos.gr> escribió:
>
>> Hi Kewin,
>>
>> i tried to force the UE to download UDP packets with throughput 70 Mbps
>> through iperf3 Reverse Mode (-R). We have a lot of losses (lost packets).
>> I have attached the UE iperf3 output. Something maybe restricts the DL
>> throughput.
>>
>> Best Regards,
>> Dimitris
>>
>> > True, but bandwith from the client side is not reliable, because it just
>> > measure the performance of your driver or hardware (as far as i know).
>> >
>> > It does not take into account re-transmissions, lost or other kind of
>> > events. With wireless communication those ones shold be taken into
>> > account.
>> >
>> > On Feb 13, 2018 7:42 PM, "Ismael Gomez" <
>> > ismael.gomez at softwareradiosystems.com> wrote:
>> >
>> >> Actually with iperf(2) you dont really need a server endpoint. UDP is
>> >> connection-less
>> >>
>> >> El mar., 13 feb. 2018 19:34, Kewin Rausch <krausch at fbk.eu> escribió:
>> >>
>> >>> Whoops, I was stuck with the Iperf(2) question.
>> >>>
>> >>> On Feb 13, 2018 7:30 PM, "Cedric Roux" <Cedric.Roux at eurecom.fr>
>> wrote:
>> >>>
>> >>>> iperf3 is different :)
>> >>>>
>> >>>> On 02/13/2018 07:06 PM, Kewin Rausch wrote:
>> >>>> > Ehm.... actually for an UDP server you should run:
>> >>>> >
>> >>>> > iperf -s -u
>> >>>> >
>> >>>> > Otherwise iperf runs a TCP server (the log on the console should
>> >>>> report it).
>> >>>> >
>> >>>> > Usually looking at the help (--help) or at online tutorials is a
>> >>>> good
>> >>>> start; try googling it.
>> >>>> >
>> >>>> > Kewin
>> >>>> >
>> >>>> > On Feb 13, 2018 6:52 PM, <dmakris at iit.demokritos.gr <mailto:
>> >>>> dmakris at iit.demokritos.gr>> wrote:
>> >>>> >
>> >>>> >     Hi Ismael,
>> >>>> >
>> >>>> >     I run the iperf client from EPC, but the Smartphone does not
>> >>>> answers, as a
>> >>>> >     server.
>> >>>> >     On the other hand, i use iperf3 server in EPC (iperf3 -s), and
>> >>>> on
>> >>>> the
>> >>>> >     smarthpone the iperf3 client(-c 192.168.3.1 -t 1000 -u -R). The
>> >>>> R,
>> >>>> means
>> >>>> >     the reverse mode. So the UE, downloads UDP packets. I set the
>> >>>> bandwidth 75
>> >>>> >     Mbps (-b 75m), but the DL throughput is only 11 Mbps.
>> >>>> >
>> >>>> >     Best regards,
>> >>>> >     Dimitris
>> >>>> >
>> >>>> >     > run the following command from EPC (amarisoft) server:
>> >>>> >     >
>> >>>> >     > iperf -c 192.168.3.2 -b75M -t100 -u
>> >>>> >     >
>> >>>> >     > to run a DL iperf and to run UL iperf:
>> >>>> >     >
>> >>>> >     > iperf -c 192.168.3.1 -b30M -t100 -u
>> >>>> >     >
>> >>>> >     >
>> >>>> >     > On Tue, 13 Feb 2018 at 10:44 <dmakris at iit.demokritos.gr
>> >>>> <mailto:
>> >>>> dmakris at iit.demokritos.gr>> wrote:
>> >>>> >     >
>> >>>> >     >> Hi Ismael,
>> >>>> >     >>
>> >>>> >     >> i use in the smartphone (client) the following:
>> >>>> >     >> -c 192.168.3.1 -t 1000 -u -R.
>> >>>> >     >> In the EPC (Amarisoft) runs the iperf -s (server).
>> >>>> >     >>
>> >>>> >     >> Best Regards,
>> >>>> >     >> Dimitris
>> >>>> >     >>
>> >>>> >     >> > what command do you use for UDP iperf?
>> >>>> >     >> >
>> >>>> >     >> > On Tue, 13 Feb 2018 at 10:34 <dmakris at iit.demokritos.gr
>> >>>> <mailto:dmakris at iit.demokritos.gr>> wrote:
>> >>>> >     >> >
>> >>>> >     >> >> Hi Ismael,
>> >>>> >     >> >>
>> >>>> >     >> >> thank you for your immediate response.
>> >>>> >     >> >> I tune the Tx and Rx gains to 88 and 45 dB respectively,
>> >>>> so
>> >>>> the snr
>> >>>> >     >> from
>> >>>> >     >> >> trace monitoring is about 27 dB.
>> >>>> >     >> >> I run the tests using speedtest from Okla, and i measure
>> >>>> the
>> >>>> DL
>> >>>> >     >> between
>> >>>> >     >> >> 12
>> >>>> >     >> >> and 20 Mbps. The UL is about 25 Mbps.
>> >>>> >     >> >> Then i run an iperf UDP test. When i try to download, the
>> >>>> DL
>> >>>> is about
>> >>>> >     >> 10
>> >>>> >     >> >> Mbps and the snr goes to -nan.
>> >>>> >     >> >>
>> >>>> >     >> >> Best Regards,
>> >>>> >     >> >> Dimitris
>> >>>> >     >> >>
>> >>>> >     >> >>
>> >>>> >     >> >> > Are you using TCP or UDP? TCP performance with srsENB
>> is
>> >>>> not quite
>> >>>> >     >> >> good
>> >>>> >     >> >> at
>> >>>> >     >> >> > the moment. Try UDP first.
>> >>>> >     >> >> >
>> >>>> >     >> >> > Also tune the TX and RX gains so that you have high SNR
>> >>>> (~27-30 dB)
>> >>>> >     >> in
>> >>>> >     >> >> > both
>> >>>> >     >> >> > directions.
>> >>>> >     >> >> >
>> >>>> >     >> >> > On Tue, 13 Feb 2018 at 10:17 <
>> dmakris at iit.demokritos.gr
>> >>>> <mailto:dmakris at iit.demokritos.gr>> wrote:
>> >>>> >     >> >> >
>> >>>> >     >> >> >> Hi srsLTE community,
>> >>>> >     >> >> >>
>> >>>> >     >> >> >>  in our testbed, we use the latest srsLTE software
>> >>>> (Latest commit
>> >>>> >     >> >> >> 2734ce6)
>> >>>> >     >> >> >> running on an Intel Core i7-7700K CPU (8x4.2 GHz),
>> with
>> >>>> 16 Gbyte
>> >>>> >     >> RAM.
>> >>>> >     >> >> >> The
>> >>>> >     >> >> >> OS is Ubuntu 16.04. For the RF front end we use the
>> >>>> Ettus
>> >>>> B210 in
>> >>>> >     >> >> SISO
>> >>>> >     >> >> >> Mode.
>> >>>> >     >> >> >>  For 1.4, 3, 5 and 10 MHz Bandwidth, we achieve the
>> >>>> expected
>> >>>> >     >> >> throughput.
>> >>>> >     >> >> >> But for 15 and 20 MHz, the DL varies between 12 and 20
>> >>>> Mbps. The
>> >>>> >     >> UL
>> >>>> >     >> >> is
>> >>>> >     >> >> >> about 25 Mbps.
>> >>>> >     >> >> >>
>> >>>> >     >> >> >>  Should we change another parameter from drb, rr, or
>> >>>> sib
>> >>>> >     >> >> configuration
>> >>>> >     >> >> >> files?
>> >>>> >     >> >> >> Thank you for your help.
>> >>>> >     >> >> >>
>> >>>> >     >> >> >> Thank you in advance,
>> >>>> >     >> >> >> Dimitris
>> >>>> >     >> >> >>
>> >>>> >     >> >> >> _______________________________________________
>> >>>> >     >> >> >> srslte-users mailing list
>> >>>> >     >> >> >> srslte-users at lists.softwareradiosystems.com <mailto:
>> >>>> srslte-users at lists.softwareradiosystems.com>
>> >>>> >     >> >> >> http://www.softwareradiosystems.com/
>> >>>> mailman/listinfo/srslte-users <http://www.softwareradiosystems.com/
>> >>>> mailman/listinfo/srslte-users>
>> >>>> >     >> >> >>
>> >>>> >     >> >> >
>> >>>> >     >> >>
>> >>>> >     >> >>
>> >>>> >     >> >>
>> >>>> >     >> >
>> >>>> >     >>
>> >>>> >     >>
>> >>>> >     >>
>> >>>> >     >
>> >>>> >
>> >>>> >
>> >>>> >     _______________________________________________
>> >>>> >     srslte-users mailing list
>> >>>> >     srslte-users at lists.softwareradiosystems.com <mailto:
>> >>>> srslte-users at lists.softwareradiosystems.com>
>> >>>> >     http://www.softwareradiosystems.com/
>> mailman/listinfo/srslte-users
>> >>>> <http://www.softwareradiosystems.com/mailman/listinfo/srslte-users>
>> >>>> >
>> >>>> >
>> >>>> >
>> >>>> > --
>> >>>> > Le informazioni contenute nella presente comunicazione sono di
>> >>>> natura privata e come tali sono da considerarsi riservate ed
>> >>>> indirizzate esclusivamente ai destinatari indicati e per le finalità
>> >>>> strettamente legate al relativo contenuto. Se avete ricevuto questo
>> >>>> messaggio per errore, vi preghiamo di eliminarlo e di inviare una
>> >>>> comunicazione all’indirizzo e-mail del mittente.
>> >>>> > --
>> >>>> > The information transmitted is intended only for the person or
>> >>>> entity
>> >>>> to which it is addressed and may contain confidential and/or
>> >>>> privileged
>> >>>> material. If you received this in error, please contact the sender
>> and
>> >>>> delete the material.
>> >>>> >
>> >>>> >
>> >>>> > _______________________________________________
>> >>>> > srslte-users mailing list
>> >>>> > srslte-users at lists.softwareradiosystems.com
>> >>>> > http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>> >>>> >
>> >>>>
>> >>>>
>> >>> --
>> >>> Le informazioni contenute nella presente comunicazione sono di natura
>> >>> privata
>> >>> e come tali sono da considerarsi riservate ed indirizzate
>> >>> esclusivamente
>> >>> ai destinatari indicati e per le finalità strettamente legate al
>> >>> relativo contenuto. Se avete ricevuto questo messaggio per errore, vi
>> >>> preghiamo di eliminarlo e di inviare una comunicazione all’indirizzo
>> >>> e-mail del mittente.
>> >>> --
>> >>> The information transmitted is intended only for the person or entity
>> >>> to
>> >>> which it is addressed and may contain confidential and/or privileged
>> >>> material. If you received this in error, please contact the sender and
>> >>> delete the material.
>> >>> _______________________________________________
>> >>> srslte-users mailing list
>> >>> srslte-users at lists.softwareradiosystems.com
>> >>> http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>> >>>
>> >>
>> >
>> > --
>> > --
>> > Le informazioni contenute nella presente comunicazione sono di natura
>> > privata
>> > e come tali sono da considerarsi riservate ed indirizzate esclusivamente
>> > ai
>> > destinatari indicati e per le finalità strettamente legate al relativo
>> > contenuto. Se avete ricevuto questo messaggio per errore, vi preghiamo
>> di
>> > eliminarlo e di inviare una comunicazione all’indirizzo e-mail del
>> > mittente.
>> > --
>> > The information transmitted is intended only for the person or entity to
>> > which it is addressed and may contain confidential and/or privileged
>> > material. If you received this in error, please contact the sender and
>> > delete the material.
>> > _______________________________________________
>> > srslte-users mailing list
>> > srslte-users at lists.softwareradiosystems.com
>> > http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>> >
>>
>

-- 
--
Le informazioni contenute nella presente comunicazione sono di natura privata 
e come tali sono da considerarsi riservate ed indirizzate esclusivamente ai 
destinatari indicati e per le finalità strettamente legate al relativo 
contenuto. Se avete ricevuto questo messaggio per errore, vi preghiamo di 
eliminarlo e di inviare una comunicazione all’indirizzo e-mail del mittente.
--
The information transmitted is intended only for the person or entity to 
which it is addressed and may contain confidential and/or privileged 
material. If you received this in error, please contact the sender and 
delete the material.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180214/656ce373/attachment.htm>


More information about the srsran-users mailing list