Ken, If you plan to use fiber you can use this MTP to 4 UPC/LC breakout cable: https://www.fs.com/products/68059.html With these generic SFP28 transceivers for the LC end: https://www.fs.com/products/75296.html
The MTP end of the connector will need to have a transceiver programmed to be compatible for your target NIC/switch. For example, targeting a Mellanox NIC you can purchase the following transceiver making sure to select the “Mellanox” compatible configuration: https://www.fs.com/products/71009.html If you want to use copper cables you can purchase this 100G to 4x25G breakout cable: https://www.fs.com/products/49617.html?attribute=10201&id=197844 When choosing cooper, make sure the configuration is the “dual compatibility” solution. This configuration will have “end A” targeting your hardware brand that the QSFP28 end will plug into (e.g., select Mellanox for a Mellanox NIC) and that “end B” is selected to be “generic”. For copper, it is also not recommended to purchase lengths longer than 3 meters. A few other notes to get the ZCU216 100G running. Jumpers need to be installed on each of the 4 headers next to the 4x SFP28 cage to enable transmit. See the ZCU216 UG for more information regarding this. The reference clock for the 100G CMAC also needs to be programmed. For this, there is a `prg_8a34001` binary distributed with the CASPER ZCU216 image in the `/home/casper/bin` directory. Hope this helps, Mitch On Jul 28, 2023, at 6:27 AM, Ken Semanov <shapkiqua...@gmail.com> wrote: Dear Mitch, Thanks for the help with this. Would you happen to know the make and model of the cable that was used for implementing the tutorial? Also the receiving NIC? I believe we have enough resources here to closely replicate your hardware. On Monday, July 24, 2023 at 7:20:45 PM UTC-4 Mitch Burnett wrote: By default, the tool will targeting the first CMAC location in this case it is the 4x25 SFP28 cage for the ZCU216. The 100G port option is the one you would want to select. Best, Mitch On Jul 24, 2023, at 3:40 PM, Ken Semanov <shapki...@gmail.com> wrote: A simulink model appears below with the hosting github. Is this tutorial intended to target a ZCU216 with a mezzanine card containing a 100GbE port, or is it intended to target to the J29 4xSFP28 cage? zcu216_tut_onehundred_gbe.slx https://github.com/mitchburnett/tutorials_devel/tree/main/rfsoc/tut_onehundred_gbe In light of the above question, what should be the port configuration on the receiving NIC? The following options are available. Thank you ! <portoptions.png> -- You received this message because you are subscribed to the Google Groups "cas...@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+un...@lists.berkeley.edu. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/83a48147-2082-41de-9095-2e15b3378e32n%40lists.berkeley.edu<https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/83a48147-2082-41de-9095-2e15b3378e32n%40lists.berkeley.edu?utm_medium=email&utm_source=footer>. <portoptions.png> -- You received this message because you are subscribed to the Google Groups "casper@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+unsubscr...@lists.berkeley.edu<mailto:casper+unsubscr...@lists.berkeley.edu>. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/6db16175-8823-458e-aa70-dcd262ef8f99n%40lists.berkeley.edu<https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/6db16175-8823-458e-aa70-dcd262ef8f99n%40lists.berkeley.edu?utm_medium=email&utm_source=footer>. -- You received this message because you are subscribed to the Google Groups "casper@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+unsubscr...@lists.berkeley.edu. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/0898C15F-E68F-45C5-92D8-138C7C2D6558%40byu.edu.
