Seriously though? Let me explain. ACID me up.
Still confused? Not sure what’s going on? Well let me show you through ACID.
Hopefully by now, you realize that I am not actually talking about the drug acid, but the ACID theory in database systems. Another hopeful thought is that you realized that I’ve been capitalizing ACID, like it’s an acronym, and if not, then I’m sorry this far into the post might have lost you already then. I first came across the acronym in reddit but after finding it, I was so entranced by that acronym that I took a tangent from some other assignments just to read up on ACID. An old and important concept in database theory, it plays a huge role in a variety of transactions, where each one has a certain characteristic. They may not necessarily pertain to the healthcare industry, but it is important to grasp this concept just as important it was to understand normalization. Image Source
Atomicity – Difficult concept, or at least the word sounds confusing, but what it means is “all or nothing”. Just like many other “A” words… well just action potential. But here’s what I mean, if an application or website is attempting to enter many records all within a transaction block, there can be a VARIETY of problems that arise that can end up corrupting the information or just running into insertion errors and only a few of those records are processed. Atomicity means that it will completely rollback the whole transaction to ensure that all of the records are either processed or none of them are processed.
Consistency – Similar to how some data warehouses work, in how it can “clean” the data to ensure that it is consistent and in a valid data scheme to be inserted into the database. Consistency ensures that the database can be brought from one valid state to another. Many DBAs will work to ensure that there is a rule, constraint, or trigger to ensure the data insertion and updates are run successfully when data is inserted, updated, removed or any any way shape or form modified.
Isolation – Jeez it took me a while to wrap my head around this because I could understand it, but I couldn’t really understand the mechanics and how it could be done to fulfill this point. However, isolation ensures that each transaction is unaware of another transaction. So an example would be a retail shoe store during Black Friday, there is the back storage room and there are the salesmen on the floor. The salesmen are selling them items very quickly leading to decreased stock in the backroom, but at the same time the back storage rooms are restocking the shoes, thus the actions of both are isolated and unaware of each other. As long as each action is continuing with the necessary time and resources, then this ensures that the salesmen will not run out of stock, and the back storage workers will always have the space to keep restocking.
Durability – What more can this mean? The data is durable, consistent and permanent once it has been completed and committed into the database unless specifically deleted by the application or other users. This part is where data warehouses can be used.
Does knowing ACID make you feel like this now?