Difference between revisions of "EJFAT Group Meeting Jan. 20, 2022"
Jump to navigation
Jump to search
Line 48: | Line 48: | ||
#** Test Plans / Goals (Mike) | #** Test Plans / Goals (Mike) | ||
#* Configuration of network switches and Host CPU for FPGA LB (Bryan) | #* Configuration of network switches and Host CPU for FPGA LB (Bryan) | ||
− | #* Status of PO for Edge to Core Test Equipment ( | + | #* Status of PO for Edge to Core Test Equipment (Amitoj) |
#* Status of PO for Network Equipment (Brent) | #* Status of PO for Network Equipment (Brent) | ||
#: | #: |
Revision as of 15:59, 24 January 2022
The meeting time is 11:00am.
Connection Info:
You can connect using ZoomGov Video conferencing (ID: 161 012 5238). (Click "Expand" to the right for details -->):
Meeting URL https://jlab-org.zoomgov.com/j/1610125238?pwd=QnEvcjV6VFFndWZsQW15SmJKU0RJZz09&from=addon Meeting ID 161 012 5238 Passcode 503371 Want to dial in from a phone? Dial one of the following numbers: US: +1 669 254 5252 or +1 646 828 7666 or +1 551 285 1373 or +1 669 216 1590 or 833 568 8864 (Toll Free) Enter the meeting ID and passcode followed by # Connecting from a room system? Dial: bjn.vc or 199.48.152.152 and enter your meeting ID & passcode
Agenda:
- Previous meeting
- Issues for Discussion:
- Status of P4 code update for packet checksum (ESnet)
- Status of new bit-file for UDP-LB U280 FPGA (ESnet)
- Status of S/W support stack for LB Host Control Plane (ESnet)
- Differences in OS Networking behaviors
- UDP AutoFragmentation
- Status of Edge to Core Test Design
- ERSAP / CLARA Data Generation for UDP-LB (Vardan)
- UDP-LB Meta Data Insertion / Packetizer (Carl)
- UDP-LB Emulator (Mike)
- UDP-Reassembly (Carl)
- Test Plans / Goals (Mike)
- Configuration of network switches and Host CPU for FPGA LB (Bryan)
- Status of PO for Edge to Core Test Equipment (Amitoj)
- Status of PO for Network Equipment (Brent)
- AOT
Minutes:
- New U280 bitfile built and currently in test
- Control Plane helper API still in development
- Must engineer complete data path from source/sink for full path MTU
- Safe choice < 1500 B
- Should engineer network for 9kB
- Layer-2 switches do not induce fragmentation
- IP headers can be designated as Do not fragment (We may want to impose this on Data Sources)
- ICMP msg feedback can indicate that device would have fragmented (but can't)
- tracepath tool can report path MTU
- LB host should send gratuitous arp msgs
- VLAN for 2 x 100Gbs switch ports for U280
- Compute Server PO: awaiting pricing for bundles/unbundled configurations
- Network PO - submitted
- Action - Amotoj - enhance EJFAT wiki page with engineering details for
- setting up 9kB MTU
- settings for Linux kernel
- Action - Bryan - engineering details for initialization of switches, hosts, FPGA for coordinated network address discovery, etc.
Attendees:
- M. S. Goodrich
- D. Lawrence
G. Heyes- Y. Kumar(ESnet)
- S. Sheldon (ESnet)
- C. Timmer
V. GyurjyanD. Abbott- B. Hess
- A. Singh
B. Morris