In our previous post , we talked about how the pre- and post-Y2K buildup of ERP systems ushered in a decade of enterprise data buildup. This was a good thing, to the extent that good data drives good decision-making. But it’s becoming a bad thing as companies fight the effects of what I call data cholesterol.
Aware that they can no longer ignore the tremendous, accelerating costs of storing superfluous data, companies are
proactively paring down their databases. They’re focusing on identifying and keeping only the data that truly meets their business objectives.
To put it another way: the mantra used to be “Data, Good.” Now it has become “Data, Good; The Right Data, Better.” (My sincere apologies to George Orwell.)
With this mantra in mind, many companies have launched enterprise data management (EDM) initiatives, usually with the help of software specifically designed for this purpose.
The Four Most Common Goals of EDM Initiatives
There’s a new generation of data management software products available from a wide range of software vendors, including IBM (Optim, formerly Princeton Softech), Informatica (formerly Applimation), HP (formerly Outerbay), and Solix. Using these and other products, companies are focusing their EDM initiatives in four areas:
1. Managing data growth. Unmanaged data growth can hamper application performance and strain your financial and technical resources, making it more difficult to complete your business-critical processes on schedule. It also stresses already-tight IT budgets, as production data proliferates through non-production copies of databases.
You can control unmanaged data growth at the source by proactively managing your enterprise application data with a simple, four-step process:
- Archive historical data to a secure archive repository.
- Maintain the archive repository as a database or as a compressed file, and store it on devices optimized for your business needs.
- Manage the archive repository over time as a corporate asset to meet regulatory and compliance objectives, as well as to provide periodic data access to meet business objectives. This step should include regularly purging unneeded data.
- Provide access to the data using corporate reporting tools, and in some cases, native application forms and reports.
2. Managing non-production (including test) data. Although most companies focus on managing test data, the truth is that there’s a need to manage all non-production data, including data used for development, training, and other purposes.
The most common way companies produce non-production data is to make copies of production data. But copying production data leads to uncontrolled investments in hardware to store and manage these (often gigantic) data sets.
Managing non-production data proactively can help deliver faster time-to-market, reduced development costs, and higher quality for both package and custom enterprise applications. To achieve these benefits, your company must be willing to invest time and effort in subsetting your production data for testing and other non-production purposes.
Using the right software tool, your organization can extract “production-quality” data that is scrubbed for compliance. This is a common method in delivery environments that have stringent privacy requirements. It also has some usefulness in production support and performance testing environments.
3. Managing data privacy. Safeguarding data privacy is not just good business practice. In many cases, it’s now the law – globally.
But with a team of developers and testers who need access to your enterprise data for development and testing, how can you protect your privacy and support compliance initiatives? The best way is to use a common data privacy technique called data masking, or obfuscation.
Data masking transforms real data into new data that’s meaningful yet concealed. Masked data retains the properties of the original data – such as width, type, context, and format – but is disguised to an extent that makes it acceptable to use in non-production environments. Even if masked data is stolen, exposed, or lost, it will be of no use to anyone.
Here’s how we recommend approaching your data privacy project:
- Understand your application and business requirements. Find out:
- Where do applications exist?
- What is the purpose of the applications?
- How closely does replacement data need to match the original data?
- How much data needs to be masked?
- Determine what you need to mask. Locate sensitive employee information such as:
- Personal data
- Bank details
- Payroll information
- Determine whether the data to be masked crosses applications and requires synchronization for accurate cross-application testing.
- Develop a masking strategy and plan to execute your requirements.
4. Managing application retirement. The need to retire enterprise applications has been around for as long as there have been enterprise applications. The traditional method of decommissioning software has been to copy applications and their associated data to tape, and to store the tapes in a corporate tape vault, never to be seen again.
But there’s a serious weakness to this approach: the longer tapes sit in a vault, the more the institutional memory of them fades. And as IT staff leave your company, they take vital application knowledge with them. Thus, when retired applications inevitably need to be found and restored on occasion, it becomes more and more difficult over time to do so.
In the current regulatory environment, it’s clear that moving retired applications to tape won’t provide adequate access to necessary information. Storing retired data and applications in an active repository is a breakthrough idea. By maintaining an active corporate repository, your company can avoid losing institutional memory of your retired applications, and manage it proactively to reduce risk at a modest cost. And, applying your favorite report writer to the repository ensures it is accessible whenever needed.
Extend Your Systems and Increase Your Agility
Organizations with production databases of 500 gigabytes or larger may benefit from implementing one or more data management solutions. By doing so, they will extend the useful life of their ERP systems and increase their agility in the no-doubt-challenging economic climate of the next decade. For more information about how to launch your EDM initiative, contact Estuate.
Read more – Estuate EDM Checklist
