[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[AWIPS #YIF-768804]: AWIPS newbie

Hi there!

Glad to virtually meet you!  And great to hear you're so familiar with other 
Unidata tools.

>  Recently I've been creating python based visualizations of
>  satellite imagery and model data. I teach synoptics at SJSU and I'm
>  thinking it might be good to expose the students to CAVE. 

I think both CAVE and python-awips could potentially be useful for you and your 
students!

Now down to your questions:

>  1) If starting from scratch and wanting to get my feet wet, do you recommend 
>  setting up CAVE client and using data from the Unidata EDEX server?

Definitely the easiest way to get started is to do exactly what you guessed: 
download a CAVE
client locally (we have direct downloads for Linux and Mac, and two options for 
Windows --
a direct download which is a bit more involved, and a VM that has some 
limitations).
Take a look at our instructions to get started there:
 http://unidata.github.io/awips2/install/install-cave/
And again, you are correct, connecting to our cloud-hosted EDEX is the quickest 
and *maybe*
best option to get started.

>  - how much slower is this than having local EDEX server? Other limitations?

This is a good question.  I'm not sure if we've directly tested performance 
against a local
machine and our cloud edex.  I will say, our cloud edex is substantially larger 
than what we
would expect most people to be running remotely, so it's probably your best 
option.
Using a local edex would allow you to control more things though: 
- if you wanted specific data we don't have but you do (maybe different models, 
or various 
other products)
- if you wanted data to stay around longer than ours (depending on the product, 
our archive
lasts from 3 to 7 days)

>  2) About how many Unidata member groups are running AWIPS/CAVE regularly
>  and how many choose to have their own EDEX server?

Another good question.  We're not exactly sure on that.  But I can tell you we 
regularly have
over 100 individual connections a day to our EDEX (meaning 100 unique machines).
And I know of at least two instances of universities and a couple private 
organizations running
their own EDEX, but I believe there are more.

>  3) How difficult is setting up and maintaining EDEX server compared with LDM 
> ingest
>  and decoding. I would consider LDM very straightforward at this point.

I would say the setup is a significant bit of work.  With our help, we could 
probably get you set
up within a few days, and there might still be a little bit of back and forth 
for some finer tuning.
The part that might take a while is acquiring the proper machine, if you don't 
already have one, that
suits what you're trying to accomplish.
Maintaining the server isn't too bad, although depending on how much data 
you're ingesting, you 
might have to monitor it fairly closely.  I wouldn't say edex crashes are a 
'regular' thing, but it
does happen, so you have to be able to stay on top of that.  Not to mention 
that datasets can change,
so you might have to make tweaks to different things such as the pqact.conf to 
make sure you're still
ingesting the data you expect to be.
Also, having prior experience with LDM would put you out in front of other 
users that we've helped
set up their own EDEX, so I think you would pick certain parts up pretty 
quickly.

>  4) do you have a recommended hardware set-up to handle CAVE duties 
> effectively? 

We have general recommendations and OS specific requirements for CAVE on our 
webpage here:
http://unidata.github.io/awips2/install/install-cave/#general-requirements

>  How about an EDEX server?

For an EDEX server, it really depends on what you're trying to do.  Our current 
configuration
uses two separate machines, although we may be expanding that in the near 
future.  We've recently
helped some users over at Texas A&M set up their own 3-machine EDEX server. 
If you want a lot of data, you will need at least one pretty powerful machine.  
If you just want
to test a dataset or two, then a smaller machine would suffice.  We can get 
into more details if 
this is something you're really interested in.

>  5) Why would one want to use Python-AWIPS instead of CAVE? More 
> customizability?

Good question!  python-awips is basically a python API to access data on an 
EDEX.  It's a 
programmatic way to access and manipulate the data.  You mentioned at the 
beginning of your email
that you're working on some python-based visualizations of satellite data.  
This is something
python-awips can possibly help with (in the case that our EDEX has satellite 
data that maybe you
can't get elsewhere, or in a nicer form).  Take a quick look at some of our 
examples on our
python-awips webpage here:
http://unidata.github.io/python-awips/examples/index.html
Specifically this one, relating to GOES:
http://unidata.github.io/python-awips/examples/generated/Satellite_Imagery.html

We also have some new GOES products that are soon to be available with our new 
AWIPS release.  
These include GeoColor, Debra Dust, Cloud Snow, Fog Probability, and a few more.

Python-awips could be a nice way to introduce your students to python 
programming.  With the 
edex server you can access all kinds of geological and geographical data like 
cities, state outlines,
CWA's, lakes, etc.  You can do most of this with cartopy as well, but I know 
cartopy doesn't have
CWA's easily available.

You can also follow the instructions for our source code with examples install 
which comes with
an environment yml and all the example notebooks found on our webpage.  The 
installation instructions
are here: 
http://unidata.github.io/python-awips/index.html#source-code-with-examples-install
 
>  Is Python-AWIPS just a wrapper around all the MetPy routines?

MetPy and python-awips are two completely separate python packages.  You can 
definitely use them
together though, and I think we do use MetPy in some of our examples.

I hope this gets you started with some answers to your first questions.  Please 
take a minute to 
download CAVE and try it out.  

Additionally, we have just recently started a blog series called AWIPS Tips 
that is released every 
other Wednesday. You can take a look at our first few blogs by going here:
https://www.unidata.ucar.edu/blogs/news/tags/awipstips

If you'd care to be in the know, feel free to join our awips mailing list (if 
you haven't already):
https://www.unidata.ucar.edu/support/index.html#mailinglists

Let me know how else we can help!

--Shay Carter
She/Her/Hers
AWIPS Software Engineer
UCAR - Unidata

If you're interested, please feel free to fill out a survey about the support 
you receive: 
https://docs.google.com/forms/d/e/1FAIpQLSeDIkdk8qUMgq8ZdM4jhP-ubJPUOr-mJMQgxInwoAWoV5QcOw/viewform

Ticket Details
===================
Ticket ID: YIF-768804
Department: Support AWIPS
Priority: Normal
Status: Open
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.