After years of working for an in-memory data-grid (IMDG) vendor, I find myself trying to correct the same set of very common clichés or misconceptions on IMDGs over and over again. Based on the facts that most IMDGs are designed in a very similar way or based on similar technology seems to only stir the fire.
Looking at the recent past and that there’s a growing need for IMDGs, I wanted to take the chance to dispel the most common myths and correct misunderstandings.
The given order below isn’t an indicator about the importance of any particular myth (all of them are equally important and wrong), but might give an indication as to how often I have heard them and why they came to my mind in that order.
1. Yet another new technology!
Probably most the common false fact; I often get told that IMDGs are yet just another new technology and that’s why you should avoid using them.
This is wrong on two different levels. Let’s start with the latter part of the claim. Even if it was new technology, we should still consider it. Our industry changes rapidly, systems need to provide access to more users each and every day, and in general technologies evolve way faster than they did 10 or 20 years ago.
So, is it just another new piece of technology? Looking at the history of IMDGs, it all began with a smallish company called Tangosol and its similarly named product in early 2000. In 2007, Tangosol was acquired by Oracle. That makes the technology almost old enough to be able to drink alcohol, at least in Germany with a legal age of 18.
Over the years the industry grew and today there are many different vendors, mainly in the Java space but also in .NET, C++ and Python. Most probably there are even more written in programming languages I’m not familiar with.