Data Management

You can't indefinitely add storage capacity to tackle the growing volume of data. You'll need to look at data management solutions

Managing data is one of the most challenging tasks for any

IT manager. One reason for this is of course its ever-growing volume, which is

the most obvious and the first target of attack. So we end up buying huge

amounts of storage capacity to store all the data. But for how long do you

actually continue buying storage capacity? There has to be a point in time when

you have to stop, look back, and assess the situation. Are you backing up the

right kind of data? Is there data that can be safely archived? How much of the

stored data is irrelevant and can be safely deleted? How frequently are your

servers running out of disk space? What's the ratio between personal and

official data on your user's machines? How much redundant data is eating into

your storage capacity? These are just a few of the questions that should be

asked when managing your organizational data.

If data is not managed



Assessing data therefore, is a much bigger challenge than planning for

storage capacity, and is unfortunately often not looked at as closely. If the

data is managed properly, you could actually reduce your purchase cycles for

storage products. Moreover, it's not as if buying additional storage devices

resolves all your problems. While it might have sufficient capacity to store all

the data, you'll face the bigger challenge of managing so many disparate

storage devices. You'll not buy the same type of storage device for all the

data. There would be tape for archiving older data, IDE drives for desktops,

SCSI or fiber channel drives for the servers or the SAN. Then you'd invest in

all the host bus adapters for connectivity. With more storage devices, you'll

be faced with the challenge of managing them. For that, you'll need to ensure

that you have the required skill sets for managing different types of devices.

You'll also need to invest in additional software for managing the various

devices. SRM or Storage Resource Management software for instance, is a

significant investment in itself. If you don't go with that, then your team

will be toiling with the management interfaces of different software. Either

way, you'll be burdened with not only managing the data, but also the devices.

So create a proper data management strategy to reduce the burden of managing so

many storage devices. Let's look at this a bit more closely. 

Know your data



Data comes from many sources. In fact, it comes in from just about

everywhere.  On one side is the

deluge of paper documents like invoices, bills, purchase orders, challans,

product blueprints, etc. Each one of these is related to a different department,

like invoices come from customers, purchase orders come from sales, challans

from dispatch, blueprints from production, and so on and so forth. Apart from

these, there's paper data that's specific to different industries. Govt.

departments store lots of records for lands, birth certificates, passports, etc.

Classifying this data is a fairly challenging task, and requires interaction

with the respective business unit heads.

Digitized data is even more complicated and spread out. In

an enterprise network, every computer has data in the form of email, various

office documents, graphics images, product blueprints, newsletters, and much

more. If we take email alone, then official email would only be a fraction of

the overall user's email. The rest would be newsletters, jokes, and personal

emails. Should you backup all that e-mail? If yes, then your storage

requirements would be pretty high. But if not, then will you establish a policy

that users can't use their official email id for personal e-mails. That's a

call that needs to be taken at a corporate level. Likewise, users store a lot of

personal data besides email on their machines. These could be songs, movies,

screensavers, small applications, etc. Are you bound to backup all that data as

well? If not, then the same policy applies here, as in e-mail. Graphics images

take up a lot of storage capacity, and not to mention network bandwidth because

of their large size. If there's a lot of graphics files being transferred over

the network, it could clog the network bandwidth. Perhaps you need to segment

the network so that users working on graphics images are on a separate subnet.

Another interesting case is when users start storing multimedia files on the

servers. Worse still, they even start playing MP3s directly off the server. This

takes a toll on the server's I/O capabilities, affecting other users accessing

it.

This is a fairly simple example from our own experience. 

The mobility factor



Nowadays, with the growing number of mobile devices, like laptops, PDAs,

smartphones, etc, the data will come from those sources as well. This raises a

number of fresh concerns. The most obvious one is security. As your employees

will carry their devices around with them and will be connecting them to various

networks, they're likely to get infected with malware. While you toil with a

new security strategy to combat that, you'll also have to work out a way of

backing up the data in those devices. What if a device gets stolen? The cost of

the device itself is miniscule compared to the value of data. Backing up laptop

data is pretty simple, as it's stored on hard drives, and connects to the

network like a desktop machine. It's the data on the smartphones and PDAs

that's going to become the concern area. Each device has its own connectivity

software for synchronizing data with a desktop. So, first you have to establish

a policy that every user does this data synchronization periodically and stores

it in a specified folder on the desktop or on the network. If you have a backup

agent running on every desktop, then that has to take this folder into account

as well. 

How do you manage it?



Just as there's so much data in so many formats and from so many sources,

there's no single solution that fits all. On one end are various document

management solutions for the data and at the other end are storage resource

management solutions for the devices. Between these, you'll also find

individual data management solutions for e-mail archival, data backup and

archival solutions. You'll have to use a mix of these solutions depending upon

your requirements.

Besides going for a solution, more challenging is to

establish data management policies and ensure that they're followed. The

solutions and policies must go hand in hand in order to be successful.