Welcome!

Apache Authors: Bob Gourley, Kevin Benedict, Jayaram Krishnaswamy, David Smith, RealWire News Distribution

Related Topics: Web 2.0, Java, SOA & WOA, Open Web, Cloud Expo

Web 2.0: Article

Maximize Your Return on Data: The New Business Imperative

Part 1: Lower the cost and complexity of sustaining your business applications

Application owners / senior-level IT allocate a certain percentage of their budget to sustaining, enhancing, and transforming their applications. In most organizations, the largest percentage of the IT spend is on sustaining the applications or basically "keep the lights on" type of activities, which leaves little money to enhance them to support business agility and new business requirements, or transform them to leap frog over the competition. Organizations that implement a solid data management strategy in support of their applications can maximize the return from their application investments. They can lower the cost of sustaining their applications, thus releasing budget to do what they should be doing, which is supporting the needs of the business. Innovative CIOs are actually rethinking their application environments, from just focusing on the application - the code, middleware, infrastructure, etc., to a new focus on the underlying data that supports the application. This new focus enables them to release budget and complexity from sustaining their apps, do what they do much better when it comes to enhancing their apps, and do brand new things they couldn't do before within their applications that will give them the competitive edge they are looking for.

How to Lower the Cost and Complexity of Sustaining Your Business Applications
Of all the things that keep organizations from realizing the best possible return on their data assets, enterprise application environments reside at the very top. Face it, application environments today are large, complex and generally inflexible constructs. Most companies have dozens of different types of key applications supporting countless business processes. Not only are the data volumes huge and the different data types numerous, but the data is often duplicative and the applications redundant across various business units. Moreover, the data is frequently hard to get to, the applications difficult to integrate and the quality of the data frequently questionable. No wonder it is such a challenge to provide a single comprehensive view of the critical data that the business uses every day.

This series of articles is aimed at helping organizations maximize the value of their data and applications, by moving beyond merely sustaining applications to enhancing them to support business agility and to transforming them to drive business innovation and growth. Because many organizations spend far too much of their IT budget on sustaining applications, it is important to first discover ways to lower the costs and complexity, freeing up budget and resources for innovation.

Cutting the Costs of "Keeping on the Lights"
There is a common set of challenges that most companies face around sustaining applications. These include:

  • Application Bloat - Whether the result of mergers and acquisitions, or business units going off and buying their own applications, many companies are rife with redundant applications that soak up maintenance time and money.
  • Data Sprawl - Companies frequently experience diminished application performance as the amount of data within the application grows. This in turn makes it difficult to meet SLAs and forces the purchase of additional hardware, leading to further costs.
  • Proliferating Integration Interfaces - There is a major challenge around integrating silo'd applications, and in dealing with the number and complexity of the interfaces required, which again increases costs.
  • Security and Privacy - Finally, there are the efforts and costs involved with securing the data in non-production applications, and the specter of fines should you fail to meet regulatory requirements around information privacy.

There are a number of initiatives that companies typically pursue to try to reduce the costs and efforts of sustaining applications. These include rationalizing the application portfolio by sending duplicative or inactive applications to the "application retirement home," archiving inactive data to improve application performance, masking sensitive data to meet security and privacy requirements, and finding ways to reduce the costs and complexity around integrating applications. The hitch is that organizations are trying to do all these things on a fixed budget and with a finite set of resources. Hence these initiatives have to be pursued very intelligently, making use of the best possible technologies to yield the greatest return on effort.

Getting the Most from Application Retirement
Capable of paying major dividends, application rationalization is an initiative being pursued by a good many organizations. Consequently, according to Gartner, by 2020, half of all applications that are running in data centers as of 2010 will be retired. If true, that represents a magnificent savings.

What does it take to retire redundant or obsolete applications and still provide seamless access to the archived data? Just because the application has outlived its usefulness, that doesn't mean that the data has. And for certain kinds of data, mandates demand that it be kept for years. Hence when an application is retired, it is increasingly necessary to archive all its data across all the data's sources.

Here is what has to happen to support successful application retirement in five steps:

  1. Mine the source metadata from the legacy application - You want to archive complete business entities, not just the transactional data but also master and reference data, and metadata.
  2. Extract and move the data - You want the ability to move, extract, and archive any data, including documents, attachments, images, and audio files associated with application and database records.
  3. Compress, secure and lockdown the archived data - You need to place it in to a secure, highly compressed, immutable file for later retrieval.
  4. Define and enforce retention policies - To ensure compliance, you need to be able to assign retention policies to different classes of archived data, apply legal holds to certain data, etc.; and to reduce the costs of managing ensured compliance, you want the ability to automatically purge expired records on a scheduled basis.
  5. Provide easy search access - You need to provide easy search and discovery access to archived data from any BI/reporting tool such as such as Crystal Reports, MicroStrategy, and Business Objects, and maintain access to archived data in database instances from existing application interfaces.

Improving Application Performance Through Archiving
The same steps, and same archiving technologies, also apply to archiving inactive data from live applications in order to improve their performance and reduce their TCO. This can take several forms, including archiving inactive data to an archive database in order to benefit from faster application response times, or archiving to an Optimized File Archive to effect substantial storage and infrastructure savings.

Importantly, a truly universal data archiving solution is strongly recommended, not only to support both application retirement and archiving from live applications, but also to ensure that you are able to leverage a single solution to address the archiving needs of all enterprise applications and databases, present and future.

Sub-setting and Masking Data in Non-Production Environments
The use of real data sets in development and test environments is widespread, and is necessary for good reasons. Frequently, this data is confidential or sensitive and subject to compliance requirements and the costs of not protecting it far outweigh the costs of doing so. Nevertheless, you need to control data management costs. Hence, when it comes to managing all the data in a test environment, you want the ability to:

  • Optimize performance and control costs by data sub-setting - Instead of using full sets of production data in test, you want the ability to create a functionally intact subset of the data, keeping only the data required by your business policies while maintaining all referential integrity. By working with a smaller set of data, you can shorten development cycles and reduce storage costs and the use of system resources.
  • Support compliance through data masking - By masking production data, you obfuscate Personally Identifiable Information and other sensitive data while preserving the data's usefulness in development and test activities.

In terms of flexibly protecting data privacy and confidentiality, Dynamic Data Masking technology can take you even further by providing real-time preventive capabilities. With this technology, flexible protection rules enable different kinds of masks to be applied dynamically to different kinds of data based on user privilege levels so you are able to engage in policy-based, selective masking and blocking of production data.

Reducing the Costs of Integrating Applications
For many organizations, much of the cost of keeping the IT lights on revolves around maintaining the "integration hairball" - the intricate web of point-to-point of interfaces between applications. According to Forrester Research, 87% of respondents to a recent IT survey indicated that they rely on hand coding for integration, and 75% of those admit that writing code for each integration effort leads to increased maintenance costs.

Another cost factor is the use of disparate integration tools so that there is no standard methodology and little economy of scale, not to mention difficulty sometimes in finding people trained in the use of a particular tool.

The way to substantially reduce the costs and complexity of integrating applications is to implement - and preferably, standardize on - a unified data integration platform with universal connectivity to data sources and targets, combined with the ability to access, transform, and integrate any data type, i.e., structured, unstructured, or semi-structured. To be fully useful, the platform also needs to support the full breadth of data latency requirements found in today's enterprises - batch, real-time, and changed data capture.

Importantly, a platform approach to integration will let you leverage a codeless development environment, so that custom-coded point-to-point interfaces and their expensive maintenance requirements become a thing of the past. Instead, development teams can leverage drag-and-drop development tools coupled with extensive reuse and sharing across projects of objects such as data mappings and transformations to speed development cycles and dramatically slash overall data integration costs.

Moving Forward Towards Enhancing Applications
The actions prescribed above have been proven to radically reduce the costs of sustaining applications, so that more resources can be applied to enhancing them and to driving innovation.

More Stories By Adam Wilson

Adam Wilson is the General Manager for Informatica’s Information Lifecycle Management Business Unit. Prior to assuming this role, he was in charge of product definition and go-to-market strategy for Informatica’s award-winning enterprise data integration platform. Mr. Wilson holds an MBA from the Kellogg School of Management and an engineering degree from Northwestern University. He can be reached at awilson@informatica.com or follow him on Twitter @ a_adam_wilson

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.