|Home||View Menu||Web Menus||Data Menu||Tools Menu||Data Centre||Services||Alerts Tool|
The QVS Data service (and Image Service) work on the basis of a timed collection. In the standard version this collects USGS, EMSC, GeoScience (Australia) and GeoNet (New Zealand) data once every five minutes on five minute boundaries, i.e. 00:05, 00:10, 00:15 etc.
In both versions you can set this to the value you want, but it cannot be less than 5 minutes or greater than 720 minutes. Grab your coffee and I will launch into the reasoning and workings.
Each minute I call a ‘tick’. The system ‘ticks’ once every minute. To determine when the time is up to go and get some data we use the function MOD. Put very simply MOD is the remainder after dividing by a specific value. For example, 501 MOD 5 = 1 since the remainder after 501 is divided by 5 is 1.
For further information visit Math Magic – Modular Math
During the course of the day there are 1440 ‘ticks’ or minutes for the system. (24 * 60). If I want to set things in motion every 5 minutes I simply divide the current number of ‘ticks’ for the day by the timing number, as in the example above. With a tick time of 5 the system will run off and collect data 288 times in a 24 hour period. Not really recommended for dial up! (See the data menu for information on how to download data when you are on dial up)
There is no ‘master’ dataset any longer in the system and each dataset is handled completely separately. There is however a master ‘provider’ but all that means is that the data from that provider gets downloaded first.
If you click on the image you can see all the combinations of ticks that result in an even number of sessions (data collections) in a full 24 hour day. For example 720 is half a day so it will fire twice. “But it only shows on one the picture?”
The reason for that is that a value of 0 always fires. For the purposes of this exercise consider there to be a 0 (zero) in there as well. 720 will fire at midnight and mid-day.
A VERY IMPORTANT POINT that you should realise is that the QVS Data Programs ALL run on UTC / Zulu / GMT. This includes the ‘ticks’ for the service. The time on your PC/Server is converted to UTC and 0 (zero) is midnight UTC thus in the above example that data collection will take place at midnight UTC and 1200 hrs UTC and NOT at YOUR midnight and mid day.
The QVS Alerts Tool ticks once every 5 minutes to check that the Data service is still ‘alive’, but only responds to new data as and when new data is downloaded. The Alerts tool looks for changes to a details file and not at the number of ticks in use by the data service.
The rest of this, and don’t worry there is not much more, relates only to the Pro version so if you are using Standard you can now relax.
USGS and EMSC update their data about every 5 minutes if there is data to update. For this reason 5 ticks is a good value for collecting that data, but other providers may not update so frequently, or just simply not have enough data to warrant collecting it every 5 minutes.
Every data job, including the main 2, has it’s own ticks in addition to the service ticks. These are added to the service ticks to produce a value for that specific job. Generally the main one will be zero thus they will fire each time the service passes a request to the collection module. (Yes they are separate entities)
You can however ‘add’ more ticks using the ticks allocated to the specific job. For example, and I will assume a standard rate of 5 for the service, if the JMO (Japan) data collection was required every 20 minutes you would simply make the JobTicks for JMO = 15
There is a small warning here. Because the job ticks are only operational when the service has ticked, job ticks should always be a multiple of the service ticks. The checking of the job ticks is only done in the data collection module so the service does not know that it may have stepped over a job tick point. The image below illustrates this.
As you can see if Chile ticks were = 3 then many of them would be missed!
For complete details of the data job settings see the topic on maintaining data jobs.
This page was last updated 17 May 2013