Get to know the Warp 10 Directory and discover functions which query it.
Warp 10 is known for its ability to handle hundreds of millions of Geo Time Series. This post will take you on a journey across the components and functions involved in the management of those series metadata.
Meet the Directory
Each Geo Time Series has attached metadata. The combination of class and labels uniquely identifies a series. There is also a secondary set of key-value pairs called attributes, which are not part of the identifier. Finally, there is a metadata tracking when a series was last updated.
The component of Warp 10 in charge of keeping track of this metadata is called the Directory. In the standalone version of Warp 10, it is included in the single process of each instance. In the distributed version of Warp 10, it is a component that can be run all by itself and can be replicated and sharded.
|Learn more about when do you need a Time Series Database|
The role of the Directory is to identify which Geo Time Series match search criteria. Those criteria are regular expressions on class and label/attributes values and optionally a criterion on the last activity timestamp. The Directory returns the list of matching Geo Time Series, which can then be used for fetching data.
A series of WarpScript functions interact with the Directory. The rest of this post will detail the usage of each of them.
FIND function will retrieve a list of Geo Time Series matching some criteria on class and labels/attributes. The result is a list of GTS with no values.
FIND function only interacts with Directory, leading to very short response time.
Note that the structure of your regular expressions may impact the performance of the Directory search. If you have regular expression criteria for alternate values (i.e. a construct of the form
A|B|C) with lots of alternatives, we encourage you to use the
REOPTALT function which will produce an optimized regular expression from a list of values.
METASET function behaves much like
FIND but will produce an object named a Metaset which can later be used as a parameter to
FETCH. A Metaset contains credentials, restrictions on timestamps and a list of Geo Time Series, all encrypted in an opaque structure.
Using a Metaset you can give access to your data without giving out your credentials (token). You can also limit the time range which can be fetched.
Lastly, for the cases where the Directory search is less performant, using a Metaset can speed up things since it contains all the GTS metadata and its use does not require a Directory search.
Sometimes you need to learn things about your Geo Time Series. The
FINDSTATS function will identify matching Geo Time Series and will return some statistics about them.
Those statistics contain estimations of the number of GTS, of classes and of label values. If the number of matching classes or labels is above some thresholds, then detailed statistics will not be returned.
This last function will identify matching Geo Time Series and return various information.
FINDSETS will analyze the matched GTS and output the list of classes and for labels and attributes a map with the list of values associated with each label or attribute.
This function is very useful for providing autocompletion in front-end applications, displaying drop-down menus with possible values for class or labels.
You now know that the Warp 10 Directory is responsible for tracking all known Geo time Series and answer search queries.
You have also discovered functions related to GTS discovery.
We hope you will get a new look at your data using these functions.
Accessing your time series data is the first thing you have to do, that is why Warp 10 offers so many options for retrieving data. Discover them in th...
New features, new version of WarpStudio, dynamic dashboard tool... Looking back over the last 12 months for the Warp 10 Platform.
In this tutorial, we will see how to aggregate by calendar duration that can be irregular (like months or years for example)
Co-Founder & Chief Technology Officer