[aprssig] Gating Objects from Internet to RF (fina?l)

Steve Dimse steve at dimse.com
Wed Jul 22 19:00:37 EDT 2009


OK, I did forget how few people drink from the full fire hose these  
days, but I also do not believe the 72GB or 98GB number, because the  
filtered streams will also be increased because these objects will  
appear in their areas. So figure this as a percent...I think 20k is  
reasonable number for the full stream. These proposed packets 3(per  
second)*103(characters per packet average)*8bits/char) or 2400 bps,  
plus overhead. So the new data is somewhere between 12 and 15% of the  
current stream, so every IGate can add at least 12% to their current  
bandwidth usage. That is a huge amount for something that few people  
will use.

Steve K4HG

On Jul 22, 2009, at 6:47 PM, Tom Hayward wrote:

> On Wed, Jul 22, 2009 at 3:31 PM, Jason KG4WSV<kg4wsv at gmail.com> wrote:
>> On Wed, Jul 22, 2009 at 4:57 PM, Tom Hayward<esarfl at gmail.com> wrote:
>>> As I count now, there are about 35 full stream
>>> clients connected to first.aprs.net,
>>
>> I counted 115, and third has a few more.  I didn't look at second.
>
> Yes, you caught me. I only counted clients connected to port 10152 the
> first time.
>
> This time, I put FIRST's status page into a spreadsheet to get some
> real data. I sorted by "Send bps". 84 clients are above 19000
> bps--these have no filter string so I assume the full feed is around
> 20000 bps.
>
> Anyway, 84 full feed clients. I'll do the math for 115 too.
>
> Lynn's text file containing all of the example objects is 204KB.
> Extreme-case, this is sent every 10 minutes (staggered, of course).
> Google says this equates to 873MB/month (for each full feed):
> http://www.google.com/search?q=%28%28204+KB+%2F+10+minutes%29%29+*+1+month
>
> Multiply that by 84 or 115 clients, and it's an added 72GB or 98GB
> respectively. Not quite 1TB.
> http://www.google.com/search?q=%28%28204+KB+%2F+10+minutes%29%29+*+1+month+*+84
>
> Again, these calculations are for the full Echolink object list every
> 10 minutes.
>
> I doubt this would cripple the Core system. I think the value-added to
> the local APRS operator is worth this cost. But that's just my
> opinion, feel free to interpret these numbers how you like.
>
> Tom KD7LXL
>
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig
>





More information about the aprssig mailing list