Originally posted by SevenandFive
reply to post by RFBurns
You're telling us they have, in essence, " one" server?


*Edit: Because I hate being the resident cynic, I'll add this disclaimer.
No, they have more than one, thats the point.
Ok lets break this down.
Lets take 1 of the siesmic sensors, for instance the BHZ sensor located at the Yellowstone lake just north of it.
Now lets say the USGS public server system comprises of 20 servers.
And lets throw in 150,000 people pulling real time data from those servers.
Lets say each of the 20 servers can serve up to 100 people simultaniously with real time data flow.
Now remember, we are pulling real time data from just 1 sensor.
Those 20 pulbic servers will start to pull that data so heavily from that 1 sensor to feed the 150,000 people requesting real time data. That puts a
tremendous strain off of that 1 poor sensor.
Suddnely that sensor detects a reading that goes byond a pre set level that the scientists have set in their system.
Would the scientists be guaranteed to get undisturbed data flow from that 1 sensor when 150,000 people are pulling data from it already?
That 1 sensor feeds two systems, theirs and the public systems. It makes perfect engineering sense to put in a preventative measure to keep that data
guaranteed from that 1 sensor to the critical systems than to appease an "ohh ahh" public yanking on the chain from that 1 sensor.
Its called "no single point failure" preventative procedure, and when you got all this geological activity going on one after another, it is darn
good that such a system is in place so that the scientists get that data, uninterrupted by "ooh ahh" lag.
The problem is not on the WAN side, it is on the local side from that 1 single line from that 1 single sensor that two systems are pulling data from
it. You might have yourself a nice t3 or greater, but calculate 150,000 people demanding real time data flow through 1 sensor via 1 line from that
sensor feeding their systems plus the public "ooh ahh" servers.
Which one would you want to get that guaranteed data?
I vote the geologists and volcanologists. My curiosity can wait to make sure those guys get that data first to let me know if there is reason to up
and run out of the way than to have that data not get to them and they miss something important that might be a clue or hint that something is on the
way.
I dont see the reason why some are not picking up on the importance of those geologists and volcanologists getting this data to check it for potential
signs of trouble comming. Our "ooh ahh" is not that important, but them having that guaranteed data flow from those individual sensors IS
important.
Perhaps folks can donate some money to put in two side by side sensors so that one can be dedicated to the scientists, and the other can be dedicated
to the "ohh ahh" and then there would be no reason to have any debate after that point..ya?
Cheers!!!!