Tuesday, July 14, 2015

Reading List on Data Storage for the Computer Illiterate

This is the second post in the boring “build a backup for your studio” series of posts. The first post is here.

The primary reason I am writing these boring posts is the occasion of having a friend of mine, a professional photographer, recover from a catastrophic data failure.  Whenever I would bring up terms like "network file server", she would put on that expression of "I am just a girl and I dont know what that means" that so many of us are so familiar with.   The good news is that even my brilliant professional photographer friend can pick up these computer terms with very little effort.

This stuff is not hard to understand.  What is hard to understand may be how things are implemented to work well, if indeed they do work well, but the basic concepts are straightforward.

The fact is that most professional users of computers, even those in their own home office or studio, will have a heterogeneous collection of files that look like they are all attached to the local computer even though they are not.  Some OS's handle this better than others "out of the box" but they all accommodate it.

Most of the time you, the user, do not care if a file is local, or on your local network, or even further afield. But you very well might care if you are your own systems administrator or your studio architect and since most of us are our own administrator, you have to know this stuff.

So get over your computer anxiety and gender bias and get this done.   Here is your Wikipedia (and one optional Dell white paper) reading list.

1. All your files on a computer is managed by a file system.

2. Most simple storage on your basic home computer is directly attached storage.

3. All modern computers today also support network attached storage.

4. Whether your storage is direct or on your local network, there are a variety of techniques designed to take these relatively cheap disks for personal use and make things more reliable. There are a variety of ways of doing this. The simplest is disk mirroring. RAID is a way of formalizing some of the existing techniques of combining multiple disks into a more reliable, or better performing, “virtual” disk.   You mostly only care about RAID 0, RAID 1 and RAID 5.

5. RAID can be implemented in hardware or software or both. People used to care passionately about which one they had, hardware or software. The reality is that you should not care which one it is as long as it is reliable, fast and low maintenance.  For those who think they care, here is a Dell white paper on the topic (optional).

6. But a file system, or a file server, or a reliable disk subsystem is not the same as having a backup system, although it may be a part of that system.

Now we can get on with the exciting yet boring design of our backup system.

No comments:

Post a Comment