Steve,
That is a big help. Is seems there is raw data missing (not present in
the rtstat records) which is used to get the average figures? (there
just isn't a record for each and every product, right?)
I'm wanting to get the latency numbers at the end of each hour from the
rtstats file of just our NEXRAD2 feed. Do you know if someone has
already developed a script to do this?
On 5/28/13 9:21 PM, Steven Emmerson wrote:
Hi Donna,
On May 28, 2013 6:54 PM, "donna Cote" <d-cote@xxxxxxxx
<mailto:d-cote@xxxxxxxx>> wrote:
>
> I need a little help understanding the values put into the rtstats files.
>
> I've named the files with the
ldmhost_ldmversion_feedtype_product_origin, so this is one NEXRAD2 file
>>
>>
>> ### my pqact entry is:
>> # rtstats-LDMVERSION/LDMHOST/FEEDTYPE/PRODUCT_ORIGIN
>> ANY ^rtstats\-(.*)\/(.*)\/(.*)\/(.*)
>> FILE -close /home/ldm/data/rtstats/\2/\3/rtstats_\2_\1_\3-\4
>
>>
rtstats_bigbird.tamu.edu_6.10.1_NEXRAD2-rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu
<http://rtstats_bigbird.tamu.edu_6.10.1_NEXRAD2-rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu>
>
>
>
> One thing I don't understand is why the "product_origin" comes across
as "rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu
<http://rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu>"
>
> Why the _v_ ?
The data-product was created on host "rpga1-phmo.nexrad.noaa.gov
<http://rpga1-phmo.nexrad.noaa.gov>" and was delivered to the host on
which rtstats(1) ran by host " idd.unidata.ucar.edu
<http://idd.unidata.ucar.edu>". Think of the "_v_" as "via".
> Next I have questions about the data. From the rtstats man page and
one example line from the file:
>
> Creation-time of most recent product
> this file has: 20130528155822
>
> Product-queue insertion-time of most recent product
> this file has: 20130528155824
>
> Receiving host (eg localhost id)
> this file has: bigbird.tamu.edu <http://bigbird.tamu.edu>
>
> Feedtype
> this file has: NEXRAD2
>
> Origin host
> this file has: rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu
<http://rpga1-phmo.nexrad.noaa.gov_v_idd.unidata.ucar.edu>
>
> Products received this hour
> this file has: 72
>
> Bytes received this hour
> this file has: 2656522
>
> Latency of most recent product
> this file has: 1.20964
>
> Average latency of products received this hour
> this file has: 1.12
>
> Peak latency@min/sec past hour
> this file has: 3@2829
>
> Version of LDM running on localhost
> this file has: 6.10.1
>
>
> I'm supposing the "Latency of most recent product" applies to the
same product that has these datetimes for "Creation-time" and
"Product-queue insertion-time". Is this latency in seconds? tenths of
seconds?
Seconds.
> And seconds also for Average latency this hour?
Seconds. Yup.
> Now this "Peak latency@min/sec past hour" throws me for a loop. Is
this latency "at" minutes "or" seconds? minutes "and" seconds? And this
is a measurement of the peak for "the past hour"? So, if an rtstats
entry has an Creation-time of 20130528161521 - that's a meager 15
minutes after the top of the hour - is this peak for the part of the
current hour? In this case, the "peak" is for the past 15 minutes? Example:
>
> 20130528161521 20130528161521 ... latency: 0.648148
> average latency: 1.14 (this is or is not the average latency since
20130528160000 ?)
> and Peak latency@min/sec past hour: 2@0318 could be read as?
> maybe: "peak latency of 2 seconds over a period of 3 mins and 18
seconds?"
> or
> "peak latency is rate of 2 seconds latency averaging over 3 mins and
18 seconds period of time?"
In your example, the peak latency is 2 seconds at 3 minutes, 18 seconds
past the hour.
> I need to explain this stuff to my boss who is very familiar with
such as latency in microseconds for packets over ethernet. Wish me luck!
>
> Thank you,
> Donna
> --
> Donna Cote
> Senior Research Associate
> The Academy for Advanced Telecommunications and Learning Technologies
> Texas A&M University
> 3139 TAMU
> College Station, Texas 77843-3139
> Office: (979) 862-3982
> Fax: (979) 862-3983
> http://academy.tamu.edu/
>
> _______________________________________________
> ldm-users mailing list
> ldm-users@xxxxxxxxxxxxxxxx <mailto:ldm-users@xxxxxxxxxxxxxxxx>
> For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
