EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  <20212022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  <20212022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: Estimate Data Rate and BW requirements
From: Mark Rivers via Tech-talk <tech-talk at aps.anl.gov>
To: "'Manoussakis, Adamandios'" <manoussakis1 at llnl.gov>
Cc: "'tech-talk at aps.anl.gov'" <tech-talk at aps.anl.gov>
Date: Wed, 10 Feb 2021 17:44:16 +0000
> However, in terms of the network bandwidth I think PVA and CA will be the same.

Unless of course one is using compression which PVA supports but CA does not.  In that case PVA can significantly reduce the required bandwidth.  For example areaDetector supports JPEG compression of NTNDArrays.  That can easily reduce the bandwidth by a factor of 10 while not significantly impacting image quality.

Mark


-----Original Message-----
From: Mark Rivers 
Sent: Wednesday, February 10, 2021 11:41 AM
To: 'Manoussakis, Adamandios' <manoussakis1 at llnl.gov>
Cc: tech-talk at aps.anl.gov
Subject: RE: Estimate Data Rate and BW requirements

> It seems like for large number of arrays PVA in epics 7 might be the better way to go?  Is just due to the way the data is stored and transferred with PVA vs CA?

There are definitely advantages to PVA vs CA for arrays: support for multi-dimensional arrays, run-time changing of the data type and array size, metadata transferred atomically with the array data, etc.  areaDetector supports transporting NTNDArrays with pvAccess, and the ImageJ viewer supports this, including compressed arrays.

However, in terms of the network bandwidth I think PVA and CA will be the same.

Mark


-----Original Message-----
From: Manoussakis, Adamandios <manoussakis1 at llnl.gov> 
Sent: Wednesday, February 10, 2021 11:36 AM
To: Mark Rivers <rivers at cars.uchicago.edu>
Cc: tech-talk at aps.anl.gov
Subject: RE: Estimate Data Rate and BW requirements

Thanks,

It seems like for large number of arrays PVA in epics 7 might be the better way to go?  Is just due to the way the data is stored and transferred with PVA vs CA?

-----Original Message-----
From: Mark Rivers <rivers at cars.uchicago.edu> 
Sent: Tuesday, February 9, 2021 1:39 PM
To: Manoussakis, Adamandios <manoussakis1 at llnl.gov>
Cc: tech-talk at aps.anl.gov
Subject: Re: Estimate Data Rate and BW requirements

Hi,


I think the bandwidth requirements depend almost entirely on the number of large arrays you will be sending via Channel Access.  If you use areaDetector and Channel Access to transport arrays from an IOC to a client viewer you can easily consume 100% of GigE bandwidth.  If you have a display with 1000 scaler PVs each changing at 10Hz that will only be 80kB/s (plus overhead), so it is less than 0.1% of GigE bandwidth.


Mark



________________________________
From: Tech-talk <tech-talk-bounces at aps.anl.gov> on behalf of Manoussakis, Adamandios via Tech-talk <tech-talk at aps.anl.gov>
Sent: Tuesday, February 9, 2021 3:04 PM
To: tech-talk at aps.anl.gov
Subject: Estimate Data Rate and BW requirements


Hello All,



Trying to understand a few things on data rate and bandwidth with regards to epics.  Most of epics is based off UDP/TCP, so I was thinking that your max will be related to TCP throughput calculations?



Is there a rule of thumb for epics on estimating BW requirements?

If the infrastructure is there to transmit and no BW issues what is the upper limit for data rate given a number of PVs?

Is there an upper limit depending on PVs and there sizes when doing transfers like caget?



Thanks

References:
Estimate Data Rate and BW requirements Manoussakis, Adamandios via Tech-talk
Re: Estimate Data Rate and BW requirements Mark Rivers via Tech-talk
RE: Estimate Data Rate and BW requirements Manoussakis, Adamandios via Tech-talk
RE: Estimate Data Rate and BW requirements Mark Rivers via Tech-talk

Navigate by Date:
Prev: RE: Estimate Data Rate and BW requirements Mark Rivers via Tech-talk
Next: Issue with record processing and being able to process records after a certain amount of delay Wang, Andrew via Tech-talk
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  <20212022  2023  2024 
Navigate by Thread:
Prev: RE: Estimate Data Rate and BW requirements Mark Rivers via Tech-talk
Next: Question about option MaxPermSize for CSS lzf neu via Tech-talk
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  <20212022  2023  2024 
ANJ, 10 Feb 2021 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·