Electronic Discovery tools that extract from backups

If your company has ever been sued, you know, just how difficult

it is to satisfy an electronic discovery request from backups.

Extracting emails and other data from backups can take months.

Even if you have the resources and relevant skills to pull it off.

Today, we talked to a company that makes doing this as simple as a handshake

while also providing a defensible audit record for your lawyers.

We're talking with Brendan Sullivan, CEO and founder of Sullivan Strickler who has

over 30 years of expertise in storage.

Learn about the challenges companies face when using legacy

and backup data for e-discovery.

The importance of defensibility and the legal process and how purpose-built

software can make all the difference.

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 and

related topics for over 30 years.

Uh, ever since I had to tell my boss.

That we had no backups of the database that we had just lost.

I don't want that to ever happen to you.

And that's why I do this podcast.

On this podcast, we turn unappreciated backup admins

and to cyber recovery heroes.

This is the backup wrap-up.

W. Curtis Preston: Welcome to the show.

I'm your host, w Curtis Preston, AKA, Mr.

Backup, and then with me, I have my emotional support consultant

as they were working on my baby.

Your, your emotional support consultant?

W. Curtis Preston: I literally was sitting right here and I had my ring

camera pointed directly on the other side of this wall at the mobile dent repair.

Individual who was doing work on my, uh, Tesla model three, for those of

you that follow the podcast, you heard, um, a week or two ago that, uh, I had a

little, a little oopsie but you know, the amazing world of the internet.

I went to Yelp and I was like, ah, I gotta find a more, you know, the, and

I, and I just, I did this thing where I, I sent it out to forbid, and then I

got, like, immediately I got this quote.

I was like, dude, I'll fix it.

I'll be there.

I'll come to your house.

I'll do it there.

He came here and it was really cool.

I could watch and I, I remember I was sh I was sending you pictures as it

was going on and you were like, oh, he is doing this, he's doing that.

He's doing all the right things.

So you were

Thank you YouTube knowledge.

W. Curtis Preston: Yeah.

Yeah.

Once again, your, your YouTube knowledge came into, came into being and he

turned it, turned out really well.

Right.

Prasanna Malaiyandi: Yeah, no, it was amazing.

And I'm surprised there aren't more mobile body shops, if you will, because if you

had to go to a body shop, you go for an estimate, then you wait, then you gotta

drop the car off, then you gotta figure out how you get to and from the place.

You wait like three days.

Actually, some Teslas you have to wait like months to get repair parts.

Right.

And then.

It's just a hassle.

And so yeah, the fact that there was a mobile guy who came out did

everything, and the work looked amazing.

So

W. Curtis Preston: You were helpful as it was going on.

I was sending photos of what was going on to you and to my wife.

You were helpful.

She was not.

Shoot.

She was like, that looks bad.

That looks bad.

And I'm like, well, of course it looks bad.

It's in the middle of like body work.

Well, it's like when you're renovating or

remodeling your house, right?

You start off with something, they tear everything down.

It looks awful, and then after they're done, you're like, wow, this is amazing.

W. Curtis Preston: yeah, yeah.

It, it, it looks yeah, like, like nothing ever happened, which is

exactly how you want it to look . So, this is a sponsored episode today.

This might be a first for me because, uh, in addition to being a sponsor,

the person that we're actually bringing on happens to be my boss.

Uh, so there's that.

He has an extensive history in data storage and retrieval.

Spanning over 30 years.

He is an electrical engineer by education with over 20 years of

experience in tape manufacturing.

After that, he spent another 20 years advising companies on what

to do with their legacy data.

He's now the CEO and founder of Sullivan Strickler, which turns

complexity into capability.

Welcome to the podcast,

Brendan Sullivan.

How's it going, Brendan?

At all.

Welcome guys.

I'm glad to be here.

W. Curtis Preston: Well, welcome, to the pod.

Let's go back to a little bit before the company.

Tell us a little bit about that.

The, the time that you spent in the manufacturing side, like how

you got to, you know, you gained the knowledge that you have now.

Um, so it's always been storage for me

in some way, shape, or form.

Um, I was, uh, at a company in South Wales that manufactured a.

60 4K static Rams in, uh, 1985.

And, um, from from there, it, the company didn't last tremendously long.

Um, and I took a job, uh, at another company that, um.

I had won a contract to manufacture the first square tape cartridge in Europe,

which was the 34 80 tape cartridge, which, uh, obviously is an IBM, uh,

mainframe tape cartridge, 200 megabytes.

And that was in bridge end in South Wales.

Um, so I took that job as a technician and I've been in tape since 1985.

And, uh.

If I last just a few more years, then I can say from start to finish, my career

will be taped in some way, shape, or form.

So yeah, so we made, uh, 34 80, um, developed into 34, not making 34 90 E.

And then also I.

Got involved and manufactured some of the TK 50, TK 52, which became

the first DLT tape cartridge run, metal metal particle coating came

in and, and, uh, we manufactured the first few versions of, uh, DLT also.

W. Curtis Preston: Yeah, for many of our listeners, those are all numbers

that just rattle around and they don't mean anything to you, but for me,

I'm thinking oh yeah, I remember.

I remember the 34 80 and the TK fifties and it, and I remember.

Then when I first saw DLT cartridge, I was like, oh, this is a, this is a TK 50.

Like, it looks the, you know, it looks physically the same, um,

just, uh, significantly different.

I can probably tell you the 27 parts that go into a DLT

or went into a DLT tape cartridge and the type of plastic that was

used on the injection molding.

Uh, but, uh, nobody, nobody is interested these days

Other than Curtis.

other than Curtis, which is why I hired.

W. Curtis Preston: Yeah.

You, you were in the business a little bit before me.

Um, just a little bit.

Uh, yeah.

I was cutting my teeth on all of those cartridges as they were.

Like getting out of the industry.

Right?

Um, like, 'cause I had nine track tapes and I had the TK fifties and I

had the, the QIC or quick cartridges.

Um, and the, the, the newest thing that we had in our world

were the, the exabyte tapes.

The a, the the eight millimeter tapes.

Right.

Uh, which then began a whole other, whole other world of tapes.

But how did you go from that to the, the consultancy world?

So, um, I ran a company, um.

That was called EMAG Solutions.

Um, I was the president and CEO for about nine years.

And, uh, what, what became clear in the early part of, uh, that tenure when I

took over, it was still very much a, a manufacturing company, but we were not.

Gonna be making the latest and greatest LTO tape cartridges and 98 forties and

all these kind of tech, uh, technologies.

They were becoming a little bit, they're open technology, but, but

in terms of, uh, development of high coercivity, um, particles that,

that, uh, that are required for, for the more modern tape storage.

We, we just didn't have the investment to do that.

So it was really a case of.

What have we learned in terms of tape manufacture, tape

design, um, tape testing?

'cause there's, as a manufacturer, as you can imagine, you're testing the quality.

What have we learned in those areas that can be applied to some kind of

services, uh, in and around them?

And, uh, it was really a case of thinking, you know, what can we do?

And so we, we actually acquired a small company, um, under

my tenure at at, at uh, emag.

Uh, called Intermedia Graphics.

And, um, and they, they'd specialized in, in, um, this, this technique called

non-Native File Restoration, which basically you restore data from somewhere

other than using the software that was originally used to create the data.

So it's, it's code that all, and originally for, for them it was codes

that maybe a Word document that was a, um.

That needed to be read from one system in another system.

And that code, the core files are the same.

The core format, data format is the same, but the way that those files are laid,

laid out on the, the storage medium are different, uh, by the backup software.

And so non-native file restoration is basically a

means to get hold of that data.

So we acquired that small company and applied it to enterprise tape

technology and also, um, middleware kind of tape technology as well.

And, and then the rest was really a little bit of luck in that, uh,

business was not tremendous, um, because it, it was hard to get that,

the knowledge of that technology out.

But we had a little bit of luck around the 2001 and to 2003

timeframe where, where eDiscovery.

Was starting to gain a foothold, uh, in the market.

And, uh, the ability, you know, the producing data, for use

of evidence in a court of law.

Um, and when it, when the data came from backup tapes, there wasn't that

many companies that were able to do it.

So we were asked on some of the big cases, you know, like, uh, Enron case

and, and uh, some other cases we were asked, can we get hold of this data?

And that led.

Our software developers to develop techniques to be able

to restore and extract and perfect and improve our software.

And, uh, by around about 2002, 2003, we were thinking, okay,

this is, this is our future.

Let's, uh, it was just growing at a, at, at a, at a rapid rate at that time.

So that's really how we got into services.

So one question, Brendan, is, I know most people who

might be listening to this podcast, they're probably like, oh, but you

have, at least in today's world, right, you have normal file servers

where you have SMB or you have NFS, and then you have object store, right?

And they're all pretty standard, open, interoperable formats.

Could you go a little into that?

Why this was such a problem in the backup space when you started this or

when you started seeing this problem?

So I'll talk generally about legacy data and um, and this is

really how the, over the last 20 years.

Um, things have evolved in that, uh, the data center typically is pushed and asked,

and Curtis knows it's only too well.

Um, I have more data.

I would like to back it up faster.

You have a backup window.

That backup window is shrinking and, um, and, you know, give me my service back.

So the backup software companies and the hardware companies, um.

We're tasked with figuring out how to move larger amounts of data from

different, uh, servers, um, to back, uh, back them up as quickly as they could.

And, and over time came things like, uh, multiplexing, multithreading,

striping, spanning, um, NAS backups, NDMP, nas filers, et cetera, et cetera.

And, um, and that's all good.

You can back up data faster and more efficiently.

But of course what you're doing with that is, is you are making the, or you

are putting the data into a storage medium in a, in a more proprietary way.

Um, and, uh, that's fine as long as you are going to be doing the restoring.

But then when the company, uh, 10 years later decides that it's gonna back up

to the cloud or decides that it's gonna back up in a different way, or they,

or they, maybe they get acquired and.

No longer, they no longer have the legato network or expert.

Now they have a Veritas expert or a a Commvault expert.

And over time you lose the ability to, to be able to restore that data and, um.

It's not always a problem, but what's also evolved in the industry is

that data has become more and more used, uh, for evidence in to prove

out legal cases and, um, privacy requirements, uh, meaning evermore.

Uh, complex, uh, challenging ways in which data has to be retained,

preserved, um, and produced.

And so the data center infrastructure is set up, let's say in 2010 for one system

that's no longer serves its purpose in 2020, except the data must be kept.

And, uh, as long as that scenario.

Um, carries on.

There will be space for companies like ours that, uh, we're

keepers of lost knowledge.

Um, we, we we're able to go back into five year, 10 year, 15, 20, 25 years and,

and produce data that, that, uh, might be required for compliance or regulation

or remediation or for business use.

Um, but largely litigation is, is a driver.

And, um, so it, it's, um.

It's, it's today's modern is tomorrow's legacy, and that's not.

Yeah.

W. Curtis Preston: Yeah, there, there's two things that, that I got out of there.

Uh, Brendan one is, and there was one that has come to my mind.

The more I work with, um, your company is that I built my own career in

doing consulting, helping people to basically redesign their backup system.

You found that there's a lot of technologies that sort of come and

go and then people lose their, you know, that, like you said, that

they lose their networker person.

What I found was even when they had a networker person, there was.

There was something about backup and, and especially tape, that it, so much of it

was counterintuitive that even if they had a networker person or a NetBackup

person, the design wasn't in keeping with how the, how, how it should be designed.

Right.

And so there was plenty of money to be made in helping people

figure out how to back up.

Right.

What, it never really occurred to me until working, um, with you, was that.

Just like there are very few people that are really good at doing backups.

There's an even smaller group of people that are really good at doing

restores the one thing about backup is that most people do it every day.

You know, if they, if they have a, a backup system, they're, they're

constantly working on their.

Backups, right?

They're working, you know, they've got a backup window.

They've gotta do backup certain, certain time.

They've gotta get 'em done in a certain si, you know, in a certain time.

And there's a new server and we gotta configure this.

So you get really good at that and you say, well, when's the

last time you did a restore?

Well, we restored this one file a month ago, uh, you know, as a test,

or maybe we do a restore of a, of, of a small file now and then to test.

But no one tests the kind of restores that.

Um, that your team is doing,

And to go along with that, Curtis, it's those restore

tests that the backup person is doing is for operational recoveries, right?

Like, Hey, I blew away a file or something.

It's not for, I need a satisfying e-discovery request, which cha its own

set of requirements that go along with it.

W. Curtis Preston: Yeah.

Which is the other, the other thing, right?

So one is people just in general aren't good at doing restores, right?

And the, the second is that, you know, you talked about

all these proprietary formats.

All of them have one thing in common, and that's that none of them were

designed for eDiscovery, right?

This is why when you and I first started talking.

I remember saying something to you of like, you do realize that I've spent

my entire career trying to talk people out of doing the thing that makes them

end up needing your services, right?

Like, please don't use your backup software as your archive

software, but no one listens to me.

Uh, you know, because the vast majority of the industry just uses

their backup system and hopes they never need, uh, to do an e-discovery.

You are exactly right.

There's, you know, I, I often, uh, say there's, there's really

two reasons for our existence.

And, uh, the, the, the first one is that, is that the

requirement to keep the data, uh.

Lasts longer than the requirement to keep the infrastructure that

you use to create the data.

That's that's point number one.

And then point number two, the requirements to produce that data.

More complex and more involved than were envision envisioned at

the time that data was created.

And of course, that's eDiscovery litigation,

compliance, regulatory reasons.

I mean, if you look at, now you know this last I.

Three, four years, uh, data privacy regulations that are, that are,

uh, taking hold, GDPR and uh, CCPA and, and, and many others.

Um, there are provisions that are, uh, being stipulated.

Things like the right to be forgotten, you know, if you've collected

personal data, then you have a right.

To be forgotten.

So you leave the company, the company has your personal data.

That might not be data that is allowed, uh, or, or they're allowed to have

in the public domain and that, so you say, okay, well I'm no longer there.

Can you remove it?

Now, if these are in backups or archives, that's a major challenge, especially if

you no longer have the infrastructure around that was used to create it.

So now.

There's a whole remediation requirement, the ability to delete on demand,

uh, from data that is pre-populated with data that you must keep.

So, um, it, it's, it's actually getting more complex, not less complex, and,

uh, which is, which is great for us.

You know, this is, this is, this is how we want them, the market to evolve.

W. Curtis Preston: Yeah, it was the same thing that happened to me, you

know, years ago when basically tape drive just kept getting faster and

faster and, um, which just made my, my side of the world worse, right?

The 'cause, the, the problem as I saw it for.

20 years was that tape drives had gotten too fast for the backup.

Right.

And the, and the, the backup can only run at one speed and the tape drive can

only run at run speed and the speeds aren't anywhere near the, the same speed.

Right.

That's how we, that's how we invented multiplexing and, you know, uh,

but the problem was that everybody always saw it as the opposite.

They're like, oh, the tape drives are slow.

And so I'm going to, um.

You know, I'm gonna buy more tape drives.

And I was always like, no.

Right.

But the faster tape drives got, the better it was for me because

the worse the backup systems were.

Um, and I just look better and better by, by fixing all those.

Right.

So you've talked about e-discovery.

What does that look like how does somebody.

Decide they need your kind of services versus whatever the alternatives might be.

I would say the vast majority of clients or prospects

that are required to produce data from backup systems, are not aware.

The type of technologies that we've developed, um, in the industry so, so

often when we get on a scoping call.

Um, you know, it might, let's just say hypothetically it might be a

NetBackup environment, and they might have to produce emails or messages

from an email archiving platform or something that's been managed

through NetBackup or backup exec.

And the conversations go, you know, and, uh, okay, so how are you gonna do this?

And, uh, it, the, the, the, and the thought process is, you know, do

we know what sessions there are?

Have you, will you do phase one imports?

Will you do phase two imports?

And we go, no, no, we don't use NetBackup we use our own software and uh, they

say, well, how are you gonna restore the data if you don't have NetBackup?

And so there's always a, um, and we'll say, we do, you know, we don't need

it because we basically trick the tape cartridge into thinking it's talking to

something like a, a NetBackup environment.

And therefore we can do all sorts of things with this data.

We don't necessarily need the tapes in sequence like the native software does.

Um, we don't.

Necessarily need to go through everything systematically to get the

complete session restored, and this is what you have to do in eDiscovery.

You have to go, the whole objective here is to cast the net wide.

Um, capture everything, um, that is relevant for the potential case and

ignore as much as you can that isn't, so you're not collecting too much data and

just go straight for the specific files.

And that, that, that leaves an ideal environment, um, for the lawyers

to be, uh, you know, somebody that can find, pinpoint, and restore

as little as possible, but restore exactly what is required and then

produce that data in a defensible way.

Preserving all metadata as you do it, and then reviewing and analyzing and, and

and producing through, through that case.

So that for us, what we've, what we've done is we've built our own software.

So we have, uh, a piece of software that we, that is fundamentally two points.

One is that it, it'll drive the devices.

So it's kinda like all the SCSI mechanisms and, and, and

what have you, all the buffers.

Um, and then the rest is we, our coders create handlers that allow us to restore.

All sorts of different backup softwares.

So whether it be NetBackup, backup exec, Commvault, Tivoli, legato,

et cetera, et cetera, et cetera.

It's the same piece of software.

And, um, what, and then, then the key skills that get added to that are, you

know, we might not need the whole session, we might not need the whole thread.

Um, so then it's a case of what, what do you need to learn about the backup

environment and what's the most cost, effective, efficient, fastest way to

go about learning in that environment?

And that's, that's really the, um, the features and benefits of our tool that,

uh, that allow us to be that early part of an eDiscovery process before the data gets

loaded into a platform where the lawyers or the paralegals can review it and decide

that it's, it's relevant for the case.

And if you compare, and I don't know if you have numbers

around this, but like doing the normal way that a backup person who was say,

using NetBackup and would've had to go through in terms of restoring each of the

sessions to a location, doing everything, using NetBackup versus using the software

that you guys have created, what sort of like efficiency gains are there to

be had by going with what you do versus like what a person would normally have

to do for a relatively large environment to handle that eDiscovery case.

Well, the, the first and most obvious one is that

we work off backups, whether or not, not whether or not that be

a disc backup or a tape backup.

We work off backups, which means that the data center folks.

Don't have to change their lives at all.

And, um, so that's a huge efficiency.

Um, hand us the data, we'll take it from here, um, that that's a,

an order of magnitude advantage.

Um, and, and also it's, it's being used to, um, the legal world.

Um, you know, they have, they have, it's all about defensibility,

all about audit trails.

All about response time, availability, project management.

Um, once you, you know, once you've been through your first a hundred projects

like that, you really get to know the kind of things that they ask for.

And that's, that's a big factor.

So a company that's going through litigation for the

first time or second time.

That's a big education they've got to go through.

So we deal with the, we deal with the backups and therefore we don't

interfere with, with normal operation.

The, the second thing is that the infrastructure that data

centers are often, um, often have, are built for purpose.

They have a certain amount of data they're backing up, they have certain processes,

and if you are going to have to succumb certain amounts of that infrastructure,

it's going to interfere with your normal.

Yeah, something's gotta give.

And then the third thing is, um, it's defensibility.

Um, you know, if, if you are in a legal situation and you're getting

accused of something, and you might ultimately have to testify in a

court of law as to what you did.

What you produced and how you produced it.

There's a, a much better defensibility from a third party company like

ours than one of your employees.

Um, that, you know, could, could potentially be argued by opposing

counsel that, oh, well how do we know that you had an interest in finding,

you know, what we're looking for?

Yeah.

So those, those three things make it, make it ideal,

really, and, and smart to, uh, pick out a third party rather than, um,

rather than actually do it yourself.

W. Curtis Preston: I think your answer was great.

Um, the, I think the question the Prasanna was maybe asking

was a little bit different.

And, um, 'cause your, your answer was about, you know, why would somebody

use you versus doing it themselves?

What I would like to see you, uh, hone in on is just the actual, let's

assume you had unlimited uh, resources.

Well, you never had unlimited resources.

But let's say you had a, you had a tape library and you had a

NetBackup server for, uh, setting aside the con, the conflicting part.

How efficient would you say your process of reading everything that needs to

be read, restoring everything needs to be restored versus doing that same

process, but using pick your favorite backup software and doing it that way.

So, um, in our data center in Atlanta, um, I

don't know how many tape drives we have, but it's probably 2000 plus.

Um, and when, if somebody drops, uh, 500 tapes on us and they don't know.

They, they, they have a keyword search requirement.

They have a file requirement, but they don't know where it is.

Um, we can go a hundred wide.

I.

Day one.

And, uh, so in terms of the speed and, and then we can just scan, uh, you know,

some of the features that I mentioned in our software, we could, it, it,

e-discovery is all about ruling out.

So it's not about throwing all of the tapes in and restoring, it's about

scanning them in the most efficient way.

Can you rule it out by date?

Can you rule it out by.

Uh, backup session type data, content type.

Um, it's all about ruling out as fast as you can.

So we go wide, fast and wide, and we rule out so.

The, the, the devil's in the detail because it depends

on what the, the matter is.

But if it was that 500 tape project that you might try and tackle internally and

you've got your NetBackup or your, your environment there, you might spend months.

Whereas we might spend days in terms of getting to, getting to the data.

So it really can be quite significant.

W. Curtis Preston: Yeah, so it sounds like it's a combination of the fact that

you actually do have, not unlimited, but a significant number of tape drive

that, you know, I've been in environments that have hundreds of tape drives,

but those are really, really rare.

Most people have just enough tape drives.

You spoke to this in your previous answer.

Most people have just enough tape drives to get the, to get the job done.

And by the way, it's not just tape driving.

We've talked a lot about tape, but they, they have a, they

have, they may have a dis array.

Um, to, to do that job and that that disc system has its job to

do, and then there's people, right?

The fact that.

You've written a piece of software that is purpose built for what you are doing.

NetBackup network or TSM, pick your favorite backup software.

None of them were written to do this.

None of them, you know, if you're looking for.

We have, we have hundreds of backups and we don't know where this file is.

They are not written for that.

Some of 'em have a little bit of search capability, but to

say, um, I've been backing up.

I mean, what percentage of of eDiscovery cases are emailed?

It's really high, right?

It is high.

Yeah, I would, it's uh, it's still very high.

It used to be a hundred percent, but it al almost a hundred percent, but

it's still still 70% I would say.

W. Curtis Preston: Yeah.

Okay.

So if you're going to do an email, e-discovery case, restoring, if

you're doing this with backups, you have to restore dozens to

hundreds of copies of exchange,

Didn't you have to do this before?

W. Curtis Preston: what's that?

Didn't you do this once for a consulting

W. Curtis Preston: Yeah, did this once.

It was very lucrative for the consulting company that I worked for at the time.

Um.

But that is it.

It's just, it's just simply not designed, you know, you talked about going wide.

Backup software isn't designed to go wide.

Backup software is designed to go narrow.

I need to restore this server, this directory, this database,

to this point in time.

It is not designed to go find needle in the haystack.

Right?

And I like what you said about how that eDiscovery is all

about, uh, what did you say?

You said all about excluding, all about

Ruling out.

W. Curtis Preston: ruling out, right?

Yeah.

Ruling out in a defensible way, right?

Here is the process by which we ruled stuff out, and, uh, if you

change your mind, we have all the metadata so we can, we can, you know.

Unru things out.

I just totally made up a phrase, but, uh, if you did that with, with,

uh, you know, uh, pick your favorite backup software every time you

go through is a separate process.

Right.

It's interesting now that I think about it because you,

you said it's about ruling out.

N that's not the way backup software works, right?

In backup software, you're, you need to tell it what you want, right?

Not, Hey, I want you to look at everything and then, but don't look at these things,

uh, and just give me these things, but give them over this, this three,

three-year period that, that just, that isn't the way backup software works.

So you've, you've raised a point that it's something

from a marketing perspective.

I've always been a little bit uncomfortable with, and it's

because of the vernacular, the terminologies that are used, backup.

So, um, uh, it's easy.

It's easier for us to say we're experts in backup.

The truth is that backup softwares, companies are experts in backup and

they, after they call the product.

BUR backup and restore, but the only get ever gets called backup.

All we do is the R.

We don't do any backup.

But if we say, but if we say we're a restore company, it can

get completely misunderstood.

So we say, yeah, we we're backup experts.

But the truth is we don't do any backup.

We only do

Well, well, and to become restore experts, you

have to be backup experts, right?

Because without that, right, you're just downstream.

So you have to really be entrenched and really deeply knowledgeable

about the backup in order to be able to do restore really well.

Uh, the one thing Curtis, uh, going back to your point about sort

of the number of tape drives and like how normally in your backup

infrastructure, you only have enough infrastructure to get the stuff done.

Imagine if you went as a backup person at a company, you went to

the CIO and was like, Hey, I need a 10 million extra dollars to have

a thousand more tape drives so I can handle an e-discovery request.

That might happen every once in a while.

Right.

Already backup folks are strapped for budget, right?

And going and saying, yes, I need this in order to be able to handle

an e-discovery request, which may or might not happen, may be unpredictable.

You don't know what the scope is.

You can't be like, Hey, I'm always gonna search these

servers, or only emails, right?

It's, it all depends on what the case is, so you can't really predict that.

And so you might have to size for the largest or size for the smallest, and

then your window goes out the door, right?

So.

W. Curtis Preston: Yeah, I do wanna speak to the the world of disc backups

because as you know, Brendan, you know, a lot of the world has gone to disc.

I mean, you guys have live in the world of tape, but you also speak the

world of disc, and so I just wanted to speak to that world as well.

Your efficiency comes from both, from the fact that you can go so wide because

you do have all those tape resources, but also because if you were handed

a, a large, uh, you know, I, I know I talked to you just today about how

a customer had shipped their, their, uh, data domain box to you, right?

And, uh, you can also very easily create multiple sessions against that.

Right.

The technology is similar.

The same but requires tweaking.

Um, our software requires tweaking.

But you know, it's interesting about that, that project that we talked

about, uh, today, that, uh, we, we were restoring exchanged instances,

thousands of custodians, um, from data domain infrastructure and, um,

the lawyers always called it tape.

And we never touched a tape.

We in the end, we, we couldn't stop them saying, you know, have

you restored these tapes yet?

Or have you restored those, those tapes yet?

We couldn't, we couldn't stop that.

So it, it was all coming off data domain.

We never touched a tape, but we still used that as.

Software and tweaked our software to be able to, uh, first restore

the, uh, um, the backups from within the data domain and then extract,

exchange, and then extract custodian, PSD messages, mailboxes, et cetera.

Um.

W. Curtis Preston: Can you speak to that word by the way?

That was a word that was new to me when coming to the company was custodian.

Um, can you speak to that word?

It's a, it's a mailbox user.

Uh, so what often gets used is if, if you are in a large company and, um.

There's 6,000, uh, mailbox users.

So you might have 6,000 mailbox users on an exchange environment,

and then three people, uh, are under investigation for whatever reason.

And we have to produce the mailboxes, multiple instances, you know,

whether they be from fos in, uh, incrementals differentials, whatever.

If we have to produce all of the instances of that mailbox user, that the term

that is used is those three custodians.

Prasanna Malaiyandi: It's a legal term, right?

Yeah, I guess.

Yeah.

W. Curtis Preston: Yeah, I, it, it's just for me, historically,

that word, I mean, it meant someone who took care of the data.

So I always thought of a data custodian as like, I have thought

of myself as a data custodian, but that's not the way it, that's not

the way it's meant in this context.

That's why I just wanted to make sure we, we threw that out.

So Brendan, I know we've been talking a lot

about e-discovery, but are there other use cases that you see people

also needing this capability?

Like I used to work for backup vendors, many backup vendors, and

one of the challenges that people would always have is like, Hey,

I have name your backup vendor.

I wanna leave them and switch to your solution, but I have all this.

Data and I don't know what to really do with it, and I don't wanna

continue paying the maintenance contracts and other things, and

I'm not sure what to do with that.

Is that something that you guys also hear from customers and do you have

the ability to help them with those

It's, uh, increasingly, increasingly, the, um, in fact, I

would, I would say that there are.

Two fundamental, um, reasons for our technology.

You know, one is the remediation side of it, um, and the other is the specific

litigation discovery side of it.

Um, so it's, it's the, the federal rules of civil procedure of, of, of, of kind,

of guide make it easy or difficult.

And they've changed over recent years that make it a little easier, a

little more justifiable to remediate.

Delete, uh, certain data.

Um, and what we think has, has, has happened from that coupled with the,

the growth in cloud and, and the demise of email archive platforms.

Um, I.

What's happened is that migration is, is much more topical and our

technologies lends itself extremely, you know, ideally for migration.

Um, but I think also, I think the, what, what so many

companies really want to do, I.

Is delete the, uh, I think I see the main objective is deletion on demand

is I think the term that I'd, I'd coin.

So, you know, if you have 50 petabytes of data, um, that spanning back 15,

20 years, the reality of you needing that or having to keep all of that.

Uh, is a nonsense.

You know, you, you don't, but there is some in there that you really

should, because if you don't, it could land you in trouble or

it could, could have some value.

And so, um, migration is the driver I.

We would like to take our legacy, uh, environment and we would like to put it in

a more modern environment, a cloud-based backup system or something like that.

And, and then we say, okay, you, you know, hundreds of terabytes

of data, we can migrate that.

And, and, and it's, it's controlled by.

Often by the bandwidth of the legacy infrastructure.

So if it's an email archiving platform, like a source one where uh, support

is going away at the end of this year, and migration of data out of those

environments, uh, are limited by the speed that that source one infrastructure

can migrate those messages out.

But you've got to.

Migrate them out to be able to remediate, to be able to delete, to be

able to keep, uh, whatever you want.

It's, it's, it's complex, it's slow.

And, um, the ability to, to take the storage environment of those legacy

environments and go fast and wide, um, is.

Is highly valuable.

Um, and, but that's the driver.

It's not just the migration, that's the driver, it's the deletion because once

you, it's like turning over the stone.

Um, this is offsite Storage vendors have made a lot of money, uh,

for many years of keeping stuff.

And if you don't look at something, uh, maybe it's not there.

If you look at it.

And you know it's there, then you have to do something with it.

So, um, the challenge is, is in, for us, in terms of what technology

we develop, we need that, uh, the clients need that technology to be

able to migrate quickly, um, delete defensively and delete on demand.

So it's, it's not just migration.

It's migration and deletion.

Um, which the term I would use.

Um, generally is remediation of, of, uh, of data.

It's a growing market and data privacy is only making that,

um, come more to the fore.

Um, and, you know, companies, there's a lot of companies out there that, that

want their, their backed up archive data in the cloud so that they can

forever remove infrastructure within their data center and just put it

all up in, in the cloud, which, um.

You know, it makes perfect sense to remediate that data

before it goes to the cloud.

Not on mass.

yeah, because even though cloud is cheap , especially

when you talk about deep storage, uh.

It is still a cost, and if you keep adding that up over the months and the

years that you're keeping that data for, right, those costs can add up.

It's only cheap.

It's only cheap on the way in.

It's, it's expensive, on the way out.

Um, egress is, is a, is is what's gonna hurt.

Although I did hear that at least AWS and some of

the other public cloud companies due to EU regulations are now,

uh, have removed egress costs if you are canceling your account.

So they're at least starting to help there, but it doesn't help if you need

to pull data out just for a one-off case or something else like that.

So.

Yeah.

W. Curtis Preston: Yeah, I, I can see all kinds of migrations here, right?

I can see, I, you know, I know that.

So there are plenty of people that are still on tape and they want to keep the

tapes, but they realize that they have a bunch of DLT one tapes and you could

put, I don't know how many DLT one tapes you can put on an LT oh eight, right?

Or an LT oh nine.

Uh, but you, but you guys can do that, right?

You can, you can migrate, you can migrate and, and remediate there.

Uh, you can keep everything.

But keep it in a much smaller space, reducing your monthly bill from whoever

it is that's holding onto your tapes.

Or if you're holding onto your own tapes, you're, you know, reduce

your data center footprint of that.

You can migrate data into some sort of cloud storage, like you

said, remediate along the way.

You say, look, we've got all these backups.

And the only thing we want to keep out of them is the exchange data.

You can do that, right?

You can, uh, extract just the exchange data or just the

whatever data, whatever it is.

Um, you can extract that out defensively because maybe they've got a, a

regulatory requirement to keep that.

Or maybe they've got a, a voice system and they have to keep those records.

Whatever it it is that, that they have to keep.

You can make sure that they keep that, but delete everything else.

Migrating that into whatever kind of system that they want to do.

Um, and I know we've even talked about, um, ways in which that

you could potentially migrate it into another backup system.

Definitely more complicated, I think.

Um, but it's doable.

But obviously it's a hard, uh, it's a high level of effort, which

would, which would be costly.

So then the customer would just have to make a, a decision as to whether or

not, I think that was not a question.

I just talked for five minutes.

Brendan, we've talked about the backups, we've talked about, uh, the

remediation side and, uh, and e-discovery.

Uh, but as I recall, there's also another part of the company

that talks about forensics.

What, um, so it sounds like you, you're just handling a, a, a, all, all

the data that needs to be collected.

Yeah, so, so, uh, the backup that we've said,

uh, is not really what we do.

We do the restore.

Um, the, it, it's the proce, the legal process is, um, best understood.

Um, from a model that's being created.

Um, it's called the EDRM model, um, electronic Discovery Reference Model.

And, um, you can follow it and there's a, there's a process which is, you know,

identification of target data over on one side and all the way to produce

in a court of law on the other side, and the restoration and production

of data from legacy environments.

We've about thus far is really on the left hand side of that ED rm,

uh, model, and more on the right.

Is, uh, the forensic side and, uh, the e-discovery side, which

is where the analysis of the data.

So just getting the target data is what we've talked about thus far.

And inevitably when we're doing that, some clients they say, well,

that's only part of the process.

Um, we'd like, um, we'd ultimately like you to actually find the data or help

us find the specific data that might.

Be used for defense or, um, for whatever, whatever the lawyers,

uh, are trying to achieve.

And so that, that part is, uh, much more analysis, keyword search, um,

artificial intelligence, machine learning, computer, uh, aided learning that allows,

um, uh, platforms to sift through.

Fully indexed, uh, all sorts of data very, very quickly and build a story

that the lawyers can use, um, as, as fast and efficiently as possible.

So we do have two other departments, um, besides the, the Restore side.

We have computer forensics where we have certified forensic examiners that

typically look at, um, you know, laptops, PCs, windows, servers, but you know, more.

More, more recently it's, it's, uh, iPhones and Androids.

Um, there's a lot of data on those devices now, um, using technologies

like Cellebrite and NK and X-rays and various other things.

So we have a computer forensics department that, uh, our examiners

specifically work in that area.

Um.

Producing and reviewing things like WhatsApp chat messages,

signal chat messages, et cetera.

Teams is coming just around the corner.

Um, and then the e-discovery platform, um, which is, um, uh, the actual,

where the data actually resides and where lawyers and paralegals will, will

search and review and then tag relevant.

Uh, data and then come back to 'em to build the stories that they will

send to opposing counsel or ultimately go, uh, go to court to produce.

Um, we also have a e-discovery department, and, um, they're not

the largest parts of our, uh, of our portfolio, but, um, they're an

essential part and they're growing.

Um, because what clients want is they want you to be able to handle everything from

the far left to as far right as possible.

Yeah, you guys are like the experts, and so if you're a

company, you have no idea where to start, and you're like, Hey, I got this request.

I have a bunch of data.

I don't know what to do.

It's like, Hey, I should give a call to you guys and be like, Hey, I need help.

Yeah.

W. Curtis Preston: Yeah, I, I think remember one of the biggest surprises

for me was when I was looking at all the things that, uh, you do,

was that e-discovery part, right?

So we were talking about Microsoft 365, and you were explaining how that you

do the, the you, you use the customers.

eDiscovery tool in Microsoft 365.

And I remember asking, well, why don't they just do that?

You're like, well, because it's really hard.

Right?

Because it's, it's a, it's a very complicated, and again, going back

to something you said much earlier, is you want to be able to, to

defensively say that you did this in the proper way, that you conducted

the search in a proper way to, so that you could say, here's what we did.

Here's how, you know, here's what we've, I'm sorry, here's what we've

collected and here's the process that we used to collect that.

And that's a defensible process.

Um, versus somebody who is clicking on Microsoft 360 five's e-discovery

button for the first time.

I could see you doing that, Curtis.

Let me tell you, Curtis is a fast learner.

W. Curtis Preston: I try, I try, I try.

Well, um, with that, uh, Brendan, thanks for coming on.

Thanks for having me.

W. Curtis Preston: And thanks persona for your usual great questions as well.

I try Curtis and Brandon.

It was nice to meet you.

You too.

You too.

W. Curtis Preston: And uh, thanks to our listeners.

Be sure to subscribe so that you don't miss an episode.

That is a wrap,