DOGE and GSA Claim $1M Savings by Ditching Old Tape Tech

you Found the backup wrap up your go-to podcast for all things

backup recovery and cyber recovery.

In this episode, we take a look at a tweet from Doge that claimed that the

GSA saved a million dollars a year by converting 14,000 magnetic tapes.

Something else.

A lot of people reached out to me and said, Hey, what did I think about this?

Is it actually possible to save that kind of money with that number of tapes?

First, let me state, we will not be actually validating the claim.

Why is that?

Well, there just isn't enough information.

We don't know what tapes they moved off of.

We don't know what kind of storage that they moved to, and without

that information, it's just simply not possible to actually validate.

We're also not going to discuss politics at all.

However, we will discuss whether or not it's even possible to save that kind of

money by modernizing your tape storage.

Plus you get an embarrassing story about me from the old days.

By the way, if you don't know who I am, I'm w Curtis Preston, AKA, Mr.

Backup, and I've been passionate about backup and recovery for

over 30 years, ever since.

I had to tell my boss that we had no backups of the production

database that we had just lost.

That's why I do this podcast.

I don't want it to happen to me.

I don't want it to happen to you.

On this podcast, we turn unappreciated backup admins into Cyber Recovery Heroes.

This is the backup wrap up.

Welcome to the show.

I. Hi, I am w Curtis Preston, AKA, Mr. Backup, and I have with me a guy that

I think might be just as excited as I am about my upcoming window repair

Prasanna Malaiyandi, how's it going?

Prasanna

I am good.

Chris, did you get your sashes?

got my, I got my, they're right here.

Look at this.

This beautiful window balance.

So for

Yeah.

who may not be aware, in addition to this being an audio podcast, we

also publish our videos on YouTube.

On YouTube.

the backup wrap up,

Yeah, you can find it.

Yeah,

see all of our expressions and how we make funny faces, especially

me when I have questions.

Um,

yeah,

you could also see Curtis's window sashes.

yeah.

No, not well.

I, I was calling East Sasha.

It turns out that's the wrong term.

This is, this is a win.

This is a window balance.

Um, basically what this is, is my window is roughly, uh, it's like three feet wide

and, um, and two and a half feet high.

It's a very big window, and it's a double pane window.

It weighs like 30 pounds.

And so without the, you know, again, with the video you can

see there's these springs.

Yeah.

Um, what this does is it, is it.

It's like the giant springs on your garage door.

It makes, it, makes it, yeah,

what was going to mind.

It's like this

yeah,

you lift up heavy weights,

yeah.

And I, I, um, it's funny, like a lot of things, I had no idea all that magic

was happening behind the scenes until, until it broke and then I couldn't

lift this window up to save my life.

Right.

Um, but no, I got it out.

I ordered the,

The

yeah.

episode after we finished recording, I think you were trying to take it out

Yeah.

almost killed yourself.

Yeah.

And I could call a guy, right?

I could very easily call the guy, but this would cost me like a

thousand bucks to have the guy come.

But this part, this part here in my hand was like, two of 'em was

like, uh, 20 bucks, you know?

So if I can fix it for 20 bucks, I'm, I'm in.

And so I knew you'd be excited about that.

if it was me, I would've just, uh, painted

guy.

no,

Oh,

over the gaps, or hired the guy to paint the gaps, so

right, yeah, yeah.

Because literally 20 feet that way, I do have a window

that the window is shattered.

It's cracked beyond repair.

And uh, so I need to have a guy.

And so it's literally, the window is like two feet by 16 inches,

and the quote I got was $600.

Wow.

the glass in there.

Yeah.

Um, anyway, yeah.

So this is why I do stuff, so, uh, here's what I wanted to talk about

and, um, I'm, I'm, I'm gonna just warn our listeners that we're, we're gonna

talk about a topic that, whichever wa, whichever side of the political.

World they sit on, they're immediately gonna get excited one way or the other.

We are not gonna do that.

Okay?

We are not gonna take a political position, um, about this topic.

We're just gonna talk about the tech.

Okay?

So.

Um, and what I'm referring to is, uh, doge had, uh, for those not in the us

Doge is this new department of government, government efficiency run by Elon Musk.

Um, and, their stated goal is to try to save the government

money and they, um, uh.

They tweeted last week.

Here's the, I'm gonna quote the tweet, the U-S-G-S-A.

So that's the Government Services Administration IT team just

saved a million dollars a year by converting 14,000 magnetic

tapes and parentheses, 70-year-old technology for information storage

to permanent modern digital records.

So a lot of people texted me and, and emailed me and, and said, you know,

Curtis, what do you think of this?

You know, and, and, and I'll be honest, my immediate reaction was incredulity.

It's like, I was like, I, I, I was like, how, how would you save,

you know, for, so first off, you know, 14,000 tapes is a lot, right?

Um, and I, and I probably my first reaction.

Was to the comment about 70-year-old information technology storage.

so I think that is referring to, so how long has tape been around?

70 years

Okay,

it was invented.

Tape was invented in the fifties, right?

Mag,

but,

rephrase.

Magnetic tape was invented in the fifties.

what we have today doesn't really look anything like what it

No.

years ago.

No.

Right.

And so, and so perhaps that was just, it, it, it certainly seemed like they were

trying to make it sound like it was old.

Yes.

Which,

Uh, yeah.

Which, which it is.

It is old.

But that's all, that's also like saying, you know, my car

is a 70-year-old technology.

Sure.

You know?

70, but, okay.

Yeah.

Yeah.

Okay.

Whatever.

Um, how long have the cars been around?

19. Oh, something.

Okay, so over a hundred years, 120 years maybe.

Um, so, uh, you know, and it's based on centuries old technology of the wheel.

yeah.

But, but so I think, right, given that this was a tweet, an

x, whatever you call it these

Right.

um, there isn't a lot of information.

So I think the first thing to call out is we don't know type of

tape technology they were using.

Right, right.

And I, but I, I, what I wanted to do is sort of evaluate this tweet

Yes.

and, and just talk about it and see is this actually

something that could be done?

Is it possible to save a million dollars a year, assuming the, um,

you know that you had 14,000 tapes.

Right.

Yep.

um, I'm gonna make a big assumption right up front, and that is, I'm.

I'm willing to bet, given that this is a government, um, department, it's one

that's been around for a while, that what they're talking about is nine track tapes.

and also to go along

Yeah.

runs on Cobolt.

This is true.

Right?

Yeah.

Thank you very much.

Yeah.

Most.

Yeah.

So again, what we're talking about is most likely mainframes running on cobol

Yeah.

mainframes often used.

Nine track tapes now.

Uh, so let for those of you, uh, you know, what, you know, for, for the, for

the video audience, I'll put up a picture of what a nine track tape looks like for

the, um, uh, for the, uh, r our role.

That's A-U-R-A-L, uh, you know, uh, listeners.

Basically, think of any old movie that you've seen computers in

Yeah.

from the sixties and seventies, even the eighties.

And think about these giant refrigerator sized tape drives that had reel to reel.

Uh, that's what a nine track tape drive looks like.

to reel, think like movie projector.

Yes.

Yes.

It, well, it's two big reels.

That, and, and each reel, and, and again, just for, just for.

Uh,

Size

know.

Yeah.

Size comparison.

A nine track tape is about the size of a medium pizza.

How's that for, for those that have, you know, just to get something in your head,

What if they, what if you don't eat pizza?

it's, who doesn't eat pizza?

Get that person out here.

That's what I'm gonna have to say.

I mean, even vegetarians eat pizza, you know?

Uh, even vegans eat pizza.

Is there a group of people that are like anti pizza?

For

No.

audience, they may not.

The sizes are different.

Okay, fine.

I don't know what to tell the international people.

But one thing though is I know we talk about mainframes and we don't know what

technologies they use at the government services agency, and so it is hard to

say are the mainframes they're running.

That old, because you could still buy a modern mainframe.

There are still companies that sell modern mainframes.

Right.

So it with modern tape drive technology and other things like that.

So

I, here, here's, here's where, here's where, but here's where I'm, here's

why I've settled on nine track tape.

We know that the Doge crew tends to be younger folk, right?

And by younger folk, I mean people younger than me, right?

So like you, you are younger folk, right?

Um.

I just think that they walked into the GSA and they saw these monstrosities and said,

oh my God, we've got to get rid of this.

And, and honestly, if I walked into a data center and saw nine track tape drives

in production, I. I would be alarmed.

Right.

And if you're not even used to tape, right.

'cause we, you know, we've talked about that, that not everybody has

even seen tape in production period.

If you walked into and you saw those things and you're like, what am I

watching a movie from the fifties?

Yeah.

So I, I just think that that was the, the, if, if they walked in and they

saw like a, a bunch of LTO drives, I don't think it would've had the effect.

That resulted in the tweet.

Yeah.

So I think for the listeners,

yeah.

are considering for this exercise that

Yeah.

they are using mainframes.

They are using nine track tapes

Right.

Yeah.

there are 14,000 of them.

14,000 of them.

Okay.

So let's talk about what that would look like.

So again, it's

tapes?

what?

Where would you store 14,000 tapes?

Well, in a tape library.

And by tape library, I don't mean.

Robot, I mean literally a library where you would put tapes.

Um, I, again, you know, I'm older than dirt, so I remember back when we had

these things in production back at the bank, and the way you store nine

track tapes is on a, it's like you've literally, they're, they're, again,

they're about the size of a medium pizza you wanna store them vertically.

You don't wanna store them.

You don't want the tape to like

I was

flatten.

Right?

Uh, you, you could theoretical.

The problem with that, like storing like a book is if they're not like

a book and they could roll out.

Hmm.

So you want, you want something to sort of hold them in place so.

Like, think about that for a minute.

So they're about an inch wide.

So a hundred tapes is, let's say 120 tapes is 120 inches, which is 10 feet, right?

So we've got 14,000 of these things.

So what is that?

That's, let's do some math here.

So

1400,

hang on.

times.

10. 1400 feet.

So 1400 feet.

So they're, they're about a, they're about, like, they're, they're a little

over, they're about 13 inches tall.

Right?

So then you're gonna have some space around that for the rack.

So let's say 18 inches, right?

So we could do like four or five high, I think.

So, um, it's gonna be, it's gonna be a pretty big, the, the

point is it's gonna be a pretty big, um, physical room to hold.

14,000 tapes.

That's number one.

The second thing is that, um, the actual drives, um, are monstrous.

It's like the size of a fridge, right.

You were mentioning

I think it's 33 inches wide.

That's a reasonably sized fridge.

And then there are also really deep, because there's all kinds

of stuff in there about, uh, like there's a vacuum, literally a vacuum

to maintain tension on the tape.

And, uh, it's just really, it, it's a very.

Complicated series of, of mechanics to, to be able to, to have this tape

do what it, what you want it to do.

So, and being a government organization, you are going to have

multiple of them because no one has made a nine track drive in years.

So you're pro, you probably have a small collection of them in

order to ha have enough running

Yeah.

Right.

Uh, because you.

I, I'm, well, there's a couple of possibilities here, right?

You, you have a, a small number and then you have a very expensive service contract

of somebody to come in and maintain them, that, that's highly possible.

it's the

Or you

likely.

Yeah, it's probably likely, right?

So they probably have like three or four, and then they have a service

contract so that at least one of them is running at all times.

Right.

Um, and, um, so just.

The physical presence of these tape drives and their associated tapes

are going to take up a significant amount of space, which has a cost.

I don't know if it approaches a million dollars a year, but again,

I've never built and maintained a government building before.

Right.

But as soon as we start talking about maintaining.

Tape drives that haven't been made for 20, 30 years, then very quickly, that could,

that could approach a million dollars.

Thing on my mind though, so there are 14,000 tapes

Uhhuh.

and assuming they're nine track, that seems like a lot of data.

But you're smiling, which

I'm,

Which means that you disagree with that statement.

yeah.

So how much data do you think, so we're gonna go, we're gonna,

we're gonna go with, again, we're assuming all kinds of things.

I'm gonna, for the purposes of discussion, assume that we're

talking about the biggest, fastest.

A tape drive that, you know, a a nine track tape drive that they made.

How much data do you suppose is on such a tape?

So given that it's probably from the seventies, eighties,

Uhhuh?

and everything was smaller, and I know we've talked about your experiences

at the bank and how you had servers that were like a hundred gigs or less.

Yeah.

Right.

I'm going to go with like 10 gigs.

That's adorable.

So at the latest available nine track tape drive, stored 150 megabytes,

Wait, what?

Gigabytes, megabytes.

a megabytes 150 megabytes.

Compressed or uncompressed.

There was no compression.

Oh,

So 150 megabytes.

Um, yeah.

so, so, okay, so that's not a lot of data.

So

So, so it's both.

It's it, it's so physically big.

The drives are really physically big.

The tapes are really physically big, physically big, but they don't,

they don't hold squat in terms of data, like 150 megabytes get, you

can't even buy like a thumb drive

So,

megabytes now.

so what are they?

So assuming that this is our backup infrastructure.

I think it's their archive infrastructure.

Okay.

That's

Yeah.

I don't think this is back.

I don't think they're making new tapes on this infrastructure.

so this is all just, which also makes sense because.

It's been

I, by the way, I could be wrong.

Yeah, yeah.

No,

I could totally be wrong.

that they are only using it for archive, but that makes sense,

Right, right.

I.

less data, less frequent data.

You just write it out and you keep it stored because you need

to put it somewhere for the

Yeah.

a hundred years.

Yeah.

Which, which brings up an interesting side issue that has nothing to do with cost.

If they wrote these, uh, this data on these tapes, and these are archived tapes.

These are archived tapes from 10, 20, 30 years ago, maybe even older.

Those tapes are not made to hold data that long.

Mm.

Right.

Modern LTO technology is, you know, uh, rated at holding data for 30 years.

This stuff there, there is, there is relatively zero chance that there

isn't bit rot on some of this data.

Right.

And that will only go, that will only

worse.

with time.

Yeah.

And so do you think that they've never touched these tapes?

I think that some of these tapes have never been touched.

Yeah.

So if you never touch the tapes and no one knows about it, do you need to keep them?

If, if bit Rod happens and no one is there to read it, did it really happen?

Exactly.

Well, that, well, you know, that is a, that is a topic that

we talk about a lot, right?

Uh, if you've got tapes.

I,

So the, the question is, do you know what's on those tapes?

Right.

Do you have a reason that I'm, I'm guessing, and I think it's

a perfectly valid assumption.

I'm assuming that, um, that they have a reason to keep

the data that's on this tape.

That this is historical GSA data, and that they have to keep it for, you know, it's

literally their purpose in life, right?

They have to be able to go back and, and, and pull up this stuff.

I just don't think they do it very often.

Yeah.

Right.

So meanwhile, and, and I'm, and I harken back, it's not the same technology,

but I do harken back to what happened when they started pulling out all

the old celluloid celluloid movies,

Hmm.

right?

So, for example, when they went to, um, I, I, I, for those of you that

are movie buffs, read the story of what happened when they went back

to do a re-release of Star Wars.

Star Wars was stored on the old celluloid stuff and it was a hot mess.

Uh, they almost didn't get it right.

, so basically the point is when you go to pull that stuff, um,

you know, also they lost the original tapes of the moon landing,

Hmm.

Well, you know

is, which is, which does not help for those who don't

think it actually happened.

But anyway, that's

else is also, people are realizing they're losing their data for,

what.

uh, apparently I think it's Warner Brother movies from the early two

Yeah, yeah, I did read about that.

Yeah.

They're no longer viewable, so.

Yeah.

So there, so, so that is a totally separate, uh, reason

why they should potentially look at migrating this old data.

But the question is, the claim was that they could save

money by doing this migration.

And most of my friends who reached out to me.

You know, had friends, you know, claims like, you know, bs.

You know, it's not true.

It couldn't, there's no possible,

so before we get into the second part, right, of

yeah,

so the tweet it said that they are using.

Permanent or what was the phrase that they used?

it, it said, um, it said permanent modern digital records.

Okay, so

I don't know what that means.

I.

does this sound like they're going to store it in some database or

know.

object store or something like that?

Whatever.

It's

Yeah, the, the, it was, there was even less information about the second half

than there was about the first half.

Okay.

But let's say whatever they pick, right?

They're gonna

Yeah.

storage system.

I, I'll just say this, I'm not even gonna try to justify or to explain

what it is that they're doing.

I'll just say if I were doing it.

This is what I would do.

would you do?

Yeah.

So first off, I would not, I would, well, there's a couple of different choices.

But I'm saying if you were already using tape and you already have a

system that knows how to talk to tape, I would probably continue to use tape.

I just would move off of the old nine track tapes and onto

something more modern like LTO.

Right.

I would probably migrate to LTO nine, which is the most recent, because we're

talking about really long-term storage.

Anyway, um, go ahead.

data can you store in an LT nine?

45 terabytes.

And you said

So,

were one 50 megabytes.

so, so, so here's the thing.

You could fit all 14,000 tapes on less than about 20% of a single LTO nine

cartridge, so they can make 10 copies.

They would've spent a thousand dollars on media.

Crazy.

Yeah.

Now I will just say this, um, it's, it is not gonna, it's not gonna be overnight.

Right.

That, that was one of the critiques that people had of the, of the tweet is

they spoke about it in the past tense.

There is zero possibility that they've done this already.

I think they just sort of figured out that, you know, they're, you know,

they're referring to it like that.

So I think what you said makes total sense.

Now, if I was dumb though, so I wanna play the other side,

Okay.

So if I was dumb and I said, okay, it's gonna be a disc

space system, doesn't matter how

Mm-hmm.

Um, it's gonna be stored on disc.

This is probably important data, which

Mm-hmm.

that I might need another copy somewhere for high availability

Mm-hmm.

Mm-hmm.

Right.

And disaster recovery.

Mm-hmm.

And then by the way, this is important data.

I don't want to lose it, which means I probably need to take a backup of it.

Right?

So now I have three additional copies.

I.

In, in, in two different things and

places

to be,

I have three additional copies on

yeah.

pieces of media, one of which is offsite, right?

Our famous, the famous 3, 2, 1 rule.

And all of this costs money, I'm probably running this in some data

center, not under someone's desk,

Right.

that it's taking power and cooling and maintenance and upkeep, patching.

Security issues of keeping everything online, all the rest of the stuff

that we talk about on this podcast.

And so they're saying that, that doing all of that cost is going to be

saving them a million dollars versus even just keeping it on nine track.

Forget about even moving to LTO.

Yeah, so, so, so here, so, so yours.

You are making an assumption there.

Yes.

That they would want to keep it on disc.

And I'm saying I don't, I don't think they have a requirement to keep it online.

I think that this is archival data and I think it's fine that it's on tape.

I would have it on multiple tapes and I would have those

tapes in multiple locations.

I do not think there's a requirement to have one of those copies on disc,

I don't think they need to, but I just, my interpretation of the tweet was they're

probably storing it on disc somewhere.

uh, already or.

They're gonna archive it to this somewhere.

I, those, those, those words, those words hurt my ears.

know.

two disc.

But, um, I'm, I'm gonna take that out, uh, because I might wanna

work for a company that does that.

But, um, so I, I just don't, I don't think that's a requirement.

Right.

And, and this is about saving money.

So I would, I would, assuming this, again, if I was running the show, I

would say, what's our requirement?

How often do we read this stuff?

Well, basically never.

So I would say, okay, well let's make sure we make it on

multiple copies, multiple tapes.

And, and, and here's, here's what's interesting is you could, even though

purchasing 14,000 backup tapes, just that is gonna cost you over a million bucks.

Yeah.

Right.

But it'd be one time purchase and then you'd be good for another 30 years.

Right.

Um, even if what you did was a one-to-one conversion of, of the be 'cause one

of the problems is even though it'd be the most inefficiently stored LTO

tape in history, right, you'd have 150 megabytes on a, assuming each of those

tapes was full, which it probably isn't the case, you'd have 150 megabytes

on a, you know, 45 terabyte tape.

It'd be ridiculous.

But one of the problems when we start talking about moving forward with

tech like this is that you don't.

You might not be able to consolidate the, the data onto

from multiple tapes onto one tape.

If you could, then that would be the way to go.

Yeah.

If you could figure out how to tell the mainframe, again, this is not, this

is not the newest technology, right?

If you could tell, if you could figure out how to tell the mainframe to take the

a hundred, you know, the 14,000 tapes.

And consolidate them onto one tape.

And then once you make that one tape, then you make multiple copies of that one tape,

then, then you, that would be amazing.

But I, I'm not sure if that's going to be possible without

a lot of specialized tech.

Which then goes back to the, to the, the cost issue.

But you're still gonna need some tech.

Even if you move it to a digital modern, I can't remember the

words, A modern digital format.

Yeah.

Right.

You're still going to need that mainframe to read the old data

Yes.

out somewhere, which means you're probably gonna need

But you,

someone who

oh, you, oh.

to do this.

will need those people, regardless of how you do it, you're gonna

need some people that know what they're talking about, right?

Like, you know, my employer, right?

You know, S two data does this all the time.

Yeah.

You're going to need somebody that knows that that's good at mainframe tapes,

you know, understands all of that stuff.

If it's the better the person you get the.

Better.

The possibility that you might actually be able to consolidate tapes

onto that would be the true goal.

If you could consolidate those 14,000 tapes onto one LTO, that would be amazing.

It's just, I'm not sure.

Again, I'm not even as old as I am.

I'm not a mainframe guy.

I've seen mainframes.

I've seen nine track drives.

I've never actually have to drive one.

Right.

So you need someone older than me, right.

We have, we have those people.

And that's two data, right?

We have one guy that's like, this is, this was his life, right?

And, and you know, uh, and he, and he's brilliant.

Um, but like, you gotta pull that guy outta retirement, you know,

pay him lots of money to, to, to figure this out and see if there's

a way to migrate those tapes.

All of that will have a cost.

Having said that.

It's still possible that you could migrate all those tapes.

You would have a one-time cost, you still would save a million dollars a year.

It's just, we're not talking about until like next year,

Yeah, I just

maybe even two years.

Okay.

Go ahead.

They just go to the local computer store or Amazon and

they buy an external hard drive.

I.

By the way, you're killing me by the way.

Let me just talk about another aspect of this.

This is gonna take a minute.

I. Yep.

Right.

So I, I ran some numbers and assuming the maximum transfer rate, assuming all of

the tape drives are full, assuming that the tape drives all work all the time.

Which, which is not a valid assumption every time I use old tape drives, you

know, they, they, they don't, they don't work, you know, they, they,

they fail on a pretty regular basis.

Um, and assuming that you could do this 24 7, which you can't, and you

And you have people there 24

have people there 24 7, which you're not gonna do that, um,

we're talking, um, it's, I think, I think it came out like 1100 hours.

Right.

When you factor in a nine to five work week, when you factor in,

let's get three running tape drives.

When you factor in the amount of project management that this is going to have to

have, because anyone who's ever done any sort of government work knows that you've

got prove and proven prove your process.

You've gotta spend a ton of time proving your process before you can go do this

because you're doing a one-way trip.

Right.

You know, to, to, to use must terms.

We're going to Mars, right?

You can, you can't.

Oh, we forgot to put the tire on.

You gotta sort it out before you've gotta do this one time, right.

Um, so you just very quickly add some, so I came up with an estimate of six

months, uh, uh, you know, of time during which you're going to be paying someone

who's three, 300 to 500 bucks an hour.

Yeah.

Uh, so, you know, you're looking at, you're looking at $300,000 or so, and, and

professional services, uh, that's just for one person and then you're gonna, they're

probably gonna need some support people.

There's a lot of moving around and moving tapes around.

Um, so this is gonna take a while.

It's not gonna.

So there's, there's my opinion, there's, there's zero chance that

this has happened already, but I don't think that makes the Tweet untrue.

and this is, I think goes back to something we talked about before.

If Bit Rod happens, notice it?

Well, that's the other thing is you won't really know if

bit rot happened, most likely

Yeah.

you go to transfer the data.

and so I, another question is maybe they just ditch the old

things and say, keep the old stuff on the old stuff moving forward.

Yeah, it's not well, well, it's, they can, the que the po, but the

point is that has a cost, right?

There's an ongoing cost.

This, this is the discussion that people get in.

There's an ongoing cost to having those tape drives.

You have to maintain them.

They have to continue to work.

Having those tape drives, having 14,000 tapes, you know, we just

talked about the fact that having.

What did I say?

A hundred?

A hundred of them having a hundred of them is 10 feet long.

okay.

know,

assume that they

there's power in, there's power in cooling for all of that.

let's just assume they're gonna keep one tape drive.

Okay.

You can't, but go ahead.

Or, okay, let's say they're gonna keep two and they're gonna keep all the tapes

Okay.

and they're not gonna touch any of the old stuff

Okay.

and they're gonna say, moving forward, we're gonna use a new form of archive.

Uh huh.

Right.

That doesn't use those old nine track tapes.

Yeah.

Okay.

Which they've probably already done all this, but go ahead.

So going back to what you had said previously or earlier.

Maintenance and upkeep and all the rest for these tape drives.

Mm-hmm.

Do you think that is where they're seeing, they're going to try to

plus managing and swapping out the tapes and everything else?

Mm-hmm.

think that is potentially where they're saying they're getting

a million dollar savings from?

I think that's exactly what I'm saying.

Okay.

I think that, I think that they're saying that the cost, the, the physical

footprint of how big these tape drives are, how big the tapes are.

Just, just, just the physical presence of them has a cost.

It's not a million dollars a year.

I, at least, I don't think, but then there's a contract, there is a service

contract to maintain those tape drives.

Um, there's also.

They got some person that looks like that, that looks like me, that

understands how this stuff works.

And tho they're not cheap.

Right?

So I think, I think that it's, it's totally true that the ex, just

the mere existence, the continued existence of these systems is

costing them a million bucks a year be because of the government aspect

yeah,

and the fact that this stuff really, really, really has to work.

And so that, that, you know, that costs a lot.

my question though, maybe I'm not stating it clearly, is.

What if you left the old stuff around and just said old stuff?

Leave it as it is.

We'll pay the sunk cost and the ongoing cost of storing things

Uh,

and

yeah, I, I, no, no, you, sorry to interrupt, but I get what you're

saying, but, and I'm just saying.

That just keeping the old stuff is costing you a million bucks a year.

If you can spend a million bucks

Hmm.

something like that, and then the, and then your million bucks a year

goes away, this is a beautiful thing.

Right.

That, that's what I'm saying.

I, I guess what I'm saying is DOGE's claim doesn't sound astronomically impossible.

I I do think it's impossible that they've done it already, but I think it's highly

possible that they're gonna do it.

Moving forward, it's gonna take them several months.

Right.

Because of the amount of project management and time and just

the, just the sheer, it's 1100 hours to read 150 megabytes off

of all tape drive, you know, off.

Yeah.

this is no different than companies that say are migrating

from LT O five to LT O nine.

Well, it's just a little bit more extreme.

a little bit more extreme, but it's

Yeah.

similar.

Like this is what

Als Also, also, sorry to interrupt you.

Zero level of automation here.

That's true.

There's no tape library for nine track drives.

You need a person standing in front of.

The tape drives to mount them, to feed the tape over, to do the thing, you

know, and put it over on the, you know, it, it, it is a very manual process.

Whereas you could absolutely buy, you could rent probably a tape

library that you could fit 14,000.

It'd be a big tape library, right?

It'd be a couple hundred feet long,

Yeah.

hundred feet long, nah, it'd be like 50 feet long to put 14,000 tapes in it.

You could push a button.

And you could come back and it would be done.

Yep.

So it's, so, it's, it's, that's why I'm saying it's, it's an extreme

version of what you're saying.

No, that's something I did not think about.

Yeah.

Was that manual effort?

Yeah.

Yeah.

Very, very manual.

are you looking for a job in standing in front of those?

Uh, tape.

drives Curtis for the

me, call me.

Well, first I'd have to learn nine track tape drives.

What's funny is I used to manage.

Nine track.

I used to manage people that mounted nine track tape drives, and I did used

to have to mount them, but I never actually had to be the one on the

keyboard doing, you know, doing the thing.

Um, and um, and, and funny story, uh, that, that figures in that.

So a hundred years ago when I was at the bank, and we had, so

we were a mainframe shop, right?

We were a credit card company.

The mainframe was in Dallas.

We were in Delaware, we had what are called channel

attached tape drives, right?

So basically the tape drives were in Delaware, the computers were in Dallas.

You could cut a tape in Delaware from the data it was in that was in Dallas.

And, um, and, and, and that, that, that was amazing to me back in the day.

Right.

The funny thing was to get data from upstairs to downstairs.

We had to cut a tape and we had to bring that tape down to downstairs, right?

And the, it was a really big operation to cut these tapes and to get them,

because this was a credit card company and we absolutely played the float.

What does that mean?

The moment your money comes in, we're gonna play with it.

We're gonna speculate on the market and all the, all the

things that a bank would do.

Play with your money.

Then, and then credit it to your account like a couple days later.

That, that's just the way it works.

Right.

If you didn't, if you weren't aware that that's happening,

that's absolutely happening.

And so it was, it was, I dunno, about a couple days, but it

was definitely some time.

And so it was a really big deal of getting these tapes downstairs.

And, uh, one year we had a mainframe outage.

Hmm.

Our first ever, I was there for almost three years and our, it was

our first ever mainframe outage.

And we were in, uh, and, and we had, my memory was that we had three

of these nine track tape drives.

It was half, was like half of the space in our tape library, tape library,

meaning the room where our tapes were.

And, um, we had a team of people whose job it was to, to mount these tapes.

It was well acknowledged that the moment the mainframe came back up,

the moment the little light came on, on the channel attached tra drives.

We were to start mounting tapes

Yeah.

because we want to play the float.

So my boss's boss's, boss's boss's boss came by.

I was already in there.

I was like, guys, you understand?

Here's the thing you see.

I want you like standing at parade rest.

You know, that's a military term, standing, you know, with your hands

behind your back standing in front of this table library watching for that light.

I heard the mainframes coming on.

The moment that light turns red, you know, put the, or probably green, I don't know,

Yeah.

whatever that light comes on.

Put the tape on, puts the, yeah, actually I think we

probably prem mounted the tapes.

The moment that light turns on, press the button, start spinning tapes and

get those tapes down to payments.

And, um, the boss's, boss's boss's boss.

Tom Thamaides was his name, and he came in there to make sure that I

knew, that I knew, so that they knew, you know, how important this was.

Right.

And so, and this was not, this did not happen every day.

We did not get

Yeah.

the, you know,

Yeah.

basically the second in command at the bank Yeah.

And is in there.

And, um, um, and so I'm like, yeah, I got it.

I got it.

Mr. Mr. Thames, I've got it.

Um, you see these guys, I literally, I am not, I literally had them

standing in front of the tape drive.

Three of them, three tape drives, three people.

We're on it.

And he's like, okay.

So he goes over and I go to let him out of the tape, let him out of the room.

And we had these, we had these mag lock, um, doors.

And the way it would work is if you, if the, if the, if the door open, we had

these, we had this piece of metal that, that would block a gap between the door

so that you wouldn't be able to open the, um, the motion detector thing.

Um, and, um, um.

So there was a button to press to open up the mag lock if, if the

door got opened wrong, and so I pressed the button to let him out.

the wrong button.

Yeah,

Which button did you push, Curtis,

the EPO button, the emergency,

not

the emergency power off,

and how

so

from that?

about three and a half hours.

Uh,

No one knew how to turn the power back on.

So I extended the first ever mainframe outage by like three hours.

Curtis did you still had a job there though, right?

I did, I did not get fired.

Uh oh.

The funniest thing is, the funniest thing is, um, my boss, Susan

Davidson, uh, great, great woman.

She was normally, she was that kind of boss that you wanted to have,

that basically, if you did something stupid, she would shield you from

Yeah.

the, like, this guy.

But in this case, this guy was standing right there

Yeah, exactly.

saw me do the thing and um, Susan Davidson came in, she came in the door and she's

like, it's, it's really quiet in here.

And, and, and she goes, um.

She said, um, what, what, what's going on?

And I'm like, um, somebody pressed the a PO button and she's this person who would

normally like protect me from such things.

Who the fuck would press the a PO button right now?

I.

Oh,

That would, that would be me.

And like, actually the guy Tom, to me, he like pointed, he like pointed to me.

oh, Curtis, do you hang your head in shame?

I did.

Uh, she did not fire me, but I did quit shortly after that.

Anyway.

So long story short, I think it's highly possible Doge may have found

a way to save a million bucks.

I think it's highly possible that this particular claim,

uh, could be, could be true.

Um, and I wish them the best of luck.

Mm-hmm.

Um, so anyway, uh, day in the life of tape drives.

Day in the life of really old tape.

How's that?

All right, well thanks for, let me like, you know, think through this, you know?

Yeah, no, it was good.

Um, learned about nine track tapes more than I ever want to.

By the way, have been around nine track tapes.

Yeah.

Yeah, my dad used to work for a company that did mainframes and every once in

a while on the weekends, I would go in with him to just hang out and he'd be

Yeah.

and he'd let me into their labs and yeah, there were nine track tapes in there.

So

So this was like in the nineties.

yeah, late

you were, how old and how old were you in then?

probably was, uh, probably eight, nine, something like that.

You're killing me smalls.

Yeah.

Oh, hope.

I hope everyone else listening feels as old as I do right now.

Uh, thanks for listening.

That is a wrap.

The backup wrap up is written, recorded, and produced by me w Curtis Preston.

If you need backup or Dr. Consulting content generation or expert witness

work, check out backup central.com.

You can also find links from my O'Reilly Books on the same website.

Remember, this is an independent podcast and any opinions that

you hear are those of the speaker and not necessarily an employer.

Thanks for listening.