NoSQL Database Design – Storing Global Variables Design Pattern

database-designdocument-databasesnosql

I'm wanting to create an admin dashboard for my app, and to show new events there. In order to know which events are new, I'll need to store at least one extra datapoint to the server: the last date at which the admin user (me) looked at the events and marked them all as read.

I'm trying to figure out where it makes sense to store that data.

I'm using Mongoose, and have a few different document types. Options I've considered include:

create an extra field on the user schema, that will be undefined for all non-admin users
create a new document type just to store admin data
use mongodb directly, rather than going through mongoose

Is one of these better than the others? Have I missed something that's obviously better than all of them?

Best Answer

I'm not familiar with Mongoose but we have started pushing configuration out of xml files and into the DB using ravendb. Our approach is to:

We create a configuration object or object graph with sensible defaults.
We write a configuration manager class that provides:
- automatic creation of the object if it does not exist
- singleton access to the class
- manages caching, etc if required

On the database level it is just a special document with a fixed ID -- hence the singleton. We have had a few approaches where we could swap various configurations into the "active" singleton as well. From the code's perspective little has changed -- we typically injected configuration using IoC containers so the codebase has been pretty agnostic. Config is just manna from heaven.

Overall I'm pretty happy with it. The biggest downside is it takes config out of version control which is a feature I've really liked for the auditability aspect.

Related Solutions

MongoDB Schema – Recommended Schema for a Quiz-Engine Scenario

I would store each guess as a separate document in a user guesses collection. The structure of each document would be as follows:

Guess
- userId
- selectedImageId
- correctImageId

You may also store more information about the corresponding quick, such as the shown images and played audio. To generate statistics, you will need to run Map/Reduce over this collection. For example, to get stats of total and correct guesses, your Map/Reduce output document structure would be:

UserStats
- userId
- guesses
- correct

The map function may look like this:

function() {
    emit(this.userId, { guesses: 1, correct: this.selectedImageId === this.correctImageId ? 1 : 0 });
}

And the reduce function:

function(key, values) {
  var result = { userId = key, guesses: 0, correct: 0 };
  values.forEach(function(value) {
    result.guesses += value.guesses;
    result.correct += value.correct;   
  });
  return result;
}

Note, on MongoDB, Map/Reduce is not run automatically as source collections are updated, so you will need to implement that on your own.

When to use wide column stores instead of document based stores

I can't answer this question for you, and no one else can either, because "Gorilla vs Shark" as noted in comments above. But I will help anyway.

You have omitted an important preceding question:

What are the characteristics of the data set I am querying?

That is just as important, if not more so, than the specific queries you want to run. Some useful questions to ask about your data are:

How much data do I have? Does it fit into memory? On one server? On a cluster?
How does my data change? Does it get mass updates on a predictable frequency? Unpredictable frequency? Does it get streamed updates? Trickled transaction updates? No updates at all?
What is the structure of the entities in my data? Are there any "one-to-many" relationships? Is it all tabular? Is it mostly tabular?
What is the sparseness of my data? Is it reasonably complete? Is it mostly empty?

If you are considering this in the abstract and don't have any specific data set in mind, then there is no reasonable answer to your question.

And even with a specific, well-defined set of data, and answers to all these questions, you still might not know without doing a bakeoff of particular implementations.

Best Answer

Related Solutions

MongoDB Schema – Recommended Schema for a Quiz-Engine Scenario

When to use wide column stores instead of document based stores

Related Topic