Hi John,
Many thanks for sharing your interesting results on processing time
returned by the wsprnet.org server!
Your recent work with WSPR-X has prodded me to think again about
possible future developments of WSPR. I'm presently leaning toward
exploring the possibility of including WSPR as a selection on the WSJT-X
"Mode" menu, so as to capitalize on the extensive rig-control
capabilities that are now part of WSJT-X. Obviously some things would
need to be added to the program -- WSPR message generator, WSPR decoder,
WSPRnet reporting, some additions to the user interface, etc. But much
of the program's structure would carry over directly for WSPR use, and
it could be easier for us to maintain one program rather than two.
(It's actually three programs now, or even four -- WSPR 2.12, WSPR 4.0,
and WSPR-X, in addition to WSJT-X.)
Comments on this possible course of action would be welcome, from anyone!
-- 73, Joe, K1JT
On 4/10/2015 3:25 PM, John C. Peterson wrote:
>
> As part of the work I have been doing on WSPR-X, I enabled some
> debugging statements. One of them saves the processing time returned by
> the wsprnet.org server when (my) "spots" are reported. Note that this
> is the time wsprnet.org reports that it used to just store the "spot"
> info in it's database, the network transit time is NOT included in the
> numbers below. I thought it would be helpful to share that info here.
>
> Anyhow, the wsprnet.org server is a bit of a victim of it's own
> success. Typical processing times are around 40-60 milliseconds, but
> on occasion can be as long as 25000 milliseconds. The changes I
> made to WSPR-X will timeout a connection after 90 seconds (although
> many of those spots still get into the database OK anyway).
>
> I have attached a PNG graphic that summarizes all the wsprnet.org
> processing times. Because the times span 3 orders of magnitude, the
> vertical axis is logarithmic scaled to see the detail. (If it does
> not make it through mailman, I will be posting the same info to
> wsprnet.org shortly which handles attachments OK).
>
> Here is a histogram of the data, which should format OK here;
>
> Summary of wsprnet.org server processing times
>
> Time Int (ms) "Raw Count" Percentage
> ------------- ----------- ----------
> [ 0, 30) 0 0.00
> [ 30, 40) 31 0.46
> [ 40, 50) 1238 18.38
> [ 50, 60) 3045 45.20
> [ 60, 70) 727 10.79
> [ 70, 80) 222 3.30
> [ 80, 90) 138 2.05
> [ 90, 100) 65 0.96
> [ 100, 120) 70 1.04
> [ 120, 140) 61 0.91
> [ 140, 160) 20 0.30
> [ 160, 180) 28 0.42
> [ 180, 200) 22 0.33
> [ 200, 300) 89 1.32
> [ 300, 400) 78 1.16
> [ 400, 500) 59 0.88
> [ 500, 600) 44 0.65
> [ 600, 700) 35 0.52
> [ 700, 800) 25 0.37
> [ 800, 900) 14 0.21
> [ 900, 1000) 16 0.24
> [ 1000, 2000) 172 2.55
> [ 2000, 3000) 116 1.72
> [ 3000, 4000) 87 1.29
> [ 4000, 5000) 85 1.26
> [ 5000, 6000) 84 1.25
> [ 6000, 7000) 67 0.99
> [ 7000, 8000) 34 0.50
> [ 8000, 9000) 14 0.21
> [ 9000,10000) 10 0.15
> [10000,15000) 16 0.24
> [15000,20000) 5 0.07
> [20000,25000) 7 0.10
> [25000, inf) 12 0.18
> -------
> Total Spots 6736
>
> wsprnet.org responded with 'I cannot connect to the database' 41 times.
------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
_______________________________________________
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
