Question: What is the grid?

Claire Lee answered on 17 Jun 2013:

Aha! So, the “grid” we as physicists speak about is a bit different to the “grid” as in electricity. I’ll tell you about the computing grid, because that’s what I know best – let me know if that’s not what you meant!

So, with all the collisions in the LHC, we get a LOT of data. Something like 3000 Terabytes per year! That’s like 600 years of mp3’s, or 4 miles of CD-ROMs stacked up.

That is a HUGE amount of data to store, and to analyse. Bear in mind that you don’t just do your analysis once on the data, you probably run your code hundreds of times. With that much data, not only would it be impossible to store by yourself for your own use, but also it would take FOREVER to run your code.

So, the grid was born.

Instead of having 3000 users, each with a copy of all the data, instead what we do is have the data stored over a bunch of computing sites all over the world. The sites are high performance computing clusters at universities, physics labs, etc, all over the place, from the US to China and Australia. Not every site has all the data. But, all the users can connect to the site, to use the data.

So instead of bringing the data to you (which would be crazy-impossible), we write our code, test it out locally to make sure it works, and then send our code to where the data is. The code runs on the clusters where the data is, and then we can download our results from the internet. Cool huh?

This whole setup is called the Grid – from the computing clusters connected al over the world, with our data stored on them, to the security system set up so that only people who are allowed to access the data can. It still takes a bit of time, but it’s way way better than trying to do it all on your own computer!