Optimize with a SATA RAID Storage Solution
Range of capacities as low as $1250 per TB. Ideal if you currently rely on servers/disks/JBODs
My company has been big into MongoDB over the past year. We've seen all kinds of MongoDB projects that we or our partners have worked on, so I figured it was worth stuffing them into a top 10 list, with the intent to enlighten those who still want to know which tasks might be best handled by the document flavor of NoSQL databases. The jobs we've encountered break down along the following lines.
Yes, LDAP is fine for identity when you're authenticating or authorizing, but what about profiling things or people that aren't strongly associated with the system? What about criminal records or child support suspects or customer rewards? What about users of promotions and what they clicked on? There's always new data to add to the user's profile, from the usual top-level stuff (phone, address, email, etc.) to information a layer below (i.e., phone type). Other database types haven't evolved fast enough to capture the hundred ways we contact each other or the dozens of ways we pay for things.
Way back when, I worked for a cell phone manufacturer (or two) and later a chemical company. Each had a weird version of the same problem: Products were composed of other products, and which products those were composed of changed over time and tended to have more than one brand or identifier. Capturing the thing that contains the thing that contains the thing is much simpler in a document database than in some other database types.
This isn't necessarily because MongoDB is a great document database, but because it has specific geospatial features. Either way, MongoDB is your friend, whether you're calculating your bike ride distance or figuring out geospecific information about your customers.
The finance industry is complicated, so don't make it more complicated than it needs to be. Investment vehicles often are composed of other investment vehicles, which are then composed of other investment vehicles. Whether this is a "bandwidth" fund or a mutual fund or a fund of funds, if you're trying to perform while flattening the data out, you may suffer. Heck, the industry is full of documents that contain documents that contain documents, so why not use a document database?
As Forrest Gump said, "it happens," and then you have lots of it. You need to categorize and say what "it" is like. MongoDB does this well. There are other database types that will also work (i.e., graph databases), but MongoDB is a fine choice.
People are social creatures, and over the last decade or so we've generated exabytes of social data. Mongo is a fine choice to handle the load. Often, people talk topically, with a lot of associated metadata. MongoDB is good for storing that too.
They don't call MongoDB a "document" database for nothing. It's great for serving up text and HTML, as well as for storing and indexing content and controlling its structure.
You have to water those flowers or serve those restaurant patrons or grow your vegetables or kill zombies or whatever. Games have goals, which consist of multiple objectives obtained through achievement or paying your way out. Whether it's a titanium rake or a BFG 9000, MongoDB can handle the concurrency and save the (often multi-level) data.