Everything you wanted to know about Document Caching but were afraid to ask

Document created by steve_posch Employee on Nov 23, 2015
Version 1Show Document
  • View in full screen mode
Everything you wanted to know about Document Caching but were afraid to ask.

In a Boomi AtomSphere process, documents flow through the steps via mapping lines.  A document normally has a predictable structure at any given point.  For example, as it goes into a mapping step, its structure will look like the source profile in the map. 

But what if you want to work with two different kinds of documents at the same time?  For example you may be working with a Purchase Order, but also want to incorporate information from an item lookup CSV.  Without the Document Caching feature, you would need to cache data in process properties which is cumbersome and memory intensive.

Feature Summary

The Document Caching feature allows you to cache documents for later use.  You can incorporate that cached data alongside the current data in your process to do joins or lookups.  You can cache any type of record (flat, xml, etc). 

A cache will have at least one index -- this is a field, or fields, that you will use later on as a key (so it is like a filter or “where” clause to get data out of your cache.)

There are two ways to add data documents to the cache. 
  • You can use the “Add to Cache” step. 
  • Or, you can automatically cache the data you pull back in an operation by configuring the “Caching” tab in the operation.
There are several ways to get data out of the cache.
  • Inside a map, right click on an appropriate node in the source profile, and the “Add Cached Data” option will appear.  Select it, and a form will appear, allowing you to incorporate your cached data.
  • You can use the “Document Cache Lookup” map function
  • Wherever you can specify a “parameter value”; e.g. when assigning a value in a set properties shape, in a decision shape, message shape, etc.
  • The data will remain in your cache for as long as your process is running.   The cache can be shared among child and parent processes (so you can write to the cache in the parent process for example, and it will still be in scope in a child process).
  • Each document will become an entry in the cache.  If you have more than one record within your document (such as multiple flat file lines) that you need to be able to pull out separately, then you will need to split before passing the data into the cache.
  • The “Add Cached Data” step will NOT filter out duplicates, so if there’s a possibility your data may have dupes it is recommended that you filter them out before passing to the “Add Cached Data” step.
  • There is currently no ability to delete or update the data once it is in the cache
For more information on how to configure the Document Cache, see Document Cache Components.
10 people found this helpful