I’m still waiting for the grade on my big final “Capstone” paper. And, no, I’m not hitting “refresh” every 30 seconds on the university’s website. I’ve managed to limit myself to checking every few hours.
In the meantime, though, my “Implementing Network Systems and Security” professor has graded all of my assignments; he hasn’t plugged the final grade into the university’s reporting system, but the numeric grades for all the assignments are there.
And…?
Because, really, what could be more awesome than Teddy Roosevelt battling Bigfoot?
98.40% overall. It could have been higher, but I dropped two questions on one of the tests. One of those questions I still dispute; the other one was just simply a matter of missing the specific answer we were looking for from the Microsoft book.
And here’s a direct quote about my paper for the class:
…This was a really well written and educational paper. It gives me ideas for future classes. Again, one of the best papers I have read.
And in a richly ironic side note, this is the very last question I had to answer for my undergraduate degree at St. Ed’s:
“Why is having a backup strategy important? Answer in your own words.”
My answer:
“Answer in your own words.” I love that. You don’t know what you’ve let yourself in for, do you?
First, some background: I work for a large computer company that has nothing to do with fruit here in the Austin area. I work in enterprise support; specifically, I work in a group that deals with backup software, tape drives and tape libraries, some magnetic storage devices, and backup appliances. Backups are what I do 40 or so hours a week. I deal with everyone from mom and pop shops to the military and other government agencies. I’ve been doing this for about five and a half years now. (I am not an official spokesperson for this company; anything I say here is my own opinion.)
To steal a line from Joel Spolsky, I don’t believe having a backup strategy is important at all. What people really need is a restore strategy. I’ve seen people call in thinking their data has been backed up and looking for help doing a restore. Only it turns out that backups hadn’t been running for months because the hardware was broken and nobody noticed. Or someone didn’t change the tape. Or they did change the tape, but the tape came apart in the drive and now is unreadable and the drive is hosed. (“Hosed” is a technical term we use in the business.) Or they didn’t make the right selection on their selection list and weren’t backing up, say, the Exchange storage group. Or there’s a subtle user interface issue in the backup software that meant they were just backing up the directory structure, and not the individual files within the directories.
So, yeah, having a good backup strategy is important because it lets you recover from hardware and software failures, because it enables compliance with legal requirements if you’re in that kind of business (“Hi. We’re the FBI. We need every email your organization has sent back to when dinosaurs walked the earth.”), because it enables disaster recovery and business continuity, and because it makes users happy if you can get back that important document or email they accidentally deleted. Among other things.
But you really need a restore strategy. Can we restore the data we need? In what time frame? How far back can we go? Do we need to stagger the restore requirements; anything in the past week can be restored in hours, anything after that may take several days while we recall tapes from offsite? What’s our budget for equipment? Do we want to adopt a two phase (magnetic disk/tape) strategy? Have we tested restores of critical data? Have we tested bare-metal restores of machines if that’s part of our strategy? If a meteor hits the building, how do we get replacement equipment? Do they still make the equipment we use? If the FBI comes in and asks for all emails related to “X”, do we have an easy way of recovering those? If not, do we want to look at some of the newer software that provides archiving and indexing capability for backup? If disaster strikes, could we run some of our existing machines as virtual machines until we got new hardware?
(Interesting fact, at least to me: the latest version of Backup Exec lets you take a backup of a physical machine and convert it straight into a virtual machine that you can just drop into a virtual environment. That strikes me as being useful as heck for disaster recovery, testing, and migration.)
Anyway, I think you get my point. A backup strategy is good. A restore strategy is better.
What the heck, let’s end with some music here.