[SRO] EJFAT LB and CODA

[Michael Goodrich]

I would like to point out that one proposed solution would use a second LB to service control/user events without the need to modify the EJFAT LB F/W.

Cheers,
-Mike

-------------------------------------------------------------
Michael Goodrich, PhD
Staff Scientist - EPSCI Group
Thomas Jefferson National Accelerator Facility
Newport News, VA
goodrich at jlab.org<mailto:%20goodrich at jlab.org>
(757) 576-7530 C

________________________________
From: SRO <sro-bounces at jlab.org> on behalf of Vardan Gyurjyan via SRO <sro at jlab.org>
Sent: Thursday, September 12, 2024 1:56 PM
To: Vardan Gyurjyan via SRO <sro at jlab.org>; David Abbott <abbottd at jlab.org>; Benjamin Raydo <braydo at jlab.org>; Bryan Moffit <moffit at jlab.org>; Ed Jastrzembski <jastrzem at jlab.org>
Cc: Yatish Kumar <yak at es.net>; Derek Howard <dhoward at es.net>; Stacey Sheldon <stac at es.net>
Subject: [SRO] EJFAT LB and CODA


Following discussions between the EJFAT developers and CODA teams, the consensus is to keep the FPGA firmware design minimal, focusing primarily on load balancing. At this stage of development, adding new features to distinguish and handle different data ID types will be deferred. Consequently, the use of EJFAT LB within the streaming CODA system will not support inbound control events. Given this direction, we propose dedicating the upcoming EPSCI SRO meeting to addressing how to define and manage streaming CODA control events without the support of EJFAT LB. Potential approaches to be discussed include handling these control events as constant metadata fields within the EVIO data structure or leveraging AFECS to orchestrate and synchronize the delivery of control events to CODA components. These topics, along with other key design decisions for handling control events and the role of CODA run-control, will form the agenda for the EPSCI SRO meeting scheduled for Monday, September 16th, at 2 PM. Please let me know if you require a formal invitation for your calendar. Additionally, I'm unsure if the ESnet team has access to Teams at JLAB. If there is interest from ESnet, I will arrange a Zoom meeting instead.



-vardan

-------------------------------------------------
Vardan H. Gyurjyan, Ph.D.
Staff Scientist
Thomas Jefferson Accelerator Facility
Newport News, VA, 23606
E-mail: gurjyan at jlab.org<mailto:gurjyan at jlab.org>
757-269-5879 (JLAB)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/sro/attachments/20240912/26b58eff/attachment.html>