Experimental Physics and
Industrial Control System

"Johnson, Andrew N. via Core-talk" <[email protected]> · Mon, 14 Jan 2019 19:22:15 +0000

On 1/14/19 11:13 AM, Michael Davidsaver via Core-talk wrote:
> On 1/14/19 8:30 AM, Ralph Lange via Core-talk wrote:
>> Make that a client that only wants to check the metadata of a huge image (e.g. to decide on the image size if it wants to read the image data itself), and it's not silly anymore.
> This seems hypothetical.  Has anyone ever wanted to do this?  What was the motivation?

Exactly that, we had an areaDetector-connected camera on a system with a
limited bandwidth, although I forget now whether it was CPU- or
network-limited. When AD was configured to serve full-frame images we
could saturate the system and cause it to drop frames by running a
second copy of our image-viewer client program, but with a suitable
region-of-interest configured it could support several viewers with no
dropped frames.

I'm not sure that we ever coded the application to not request the image
data when it was full-frame so you could still discount this as another
hypothetical example, especially as the AreaDetector PVA server plugin
uses pvDatabase to serve its NTNDArray channels so is not affected by
QSRV's different behaviour. My point is that users are growing
accustomed to being able to request a subset of the available fields
from AD and other pvDatabase-based PVA servers, since they follow
behaviour which has always been documented and IIRC was covered in
previous V4 training materials. I'm looking forward to our never having
to explain again why QSRV behaves differently.

- Andrew

-- 
Arguing for surveillance because you have nothing to hide is no
different than making the claim, "I don't care about freedom of
speech because I have nothing to say." -- Edward Snowdon

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System