Shahid N. Shah's blog

Federal agencies are facing critical needs for information technology upgrades and enhancements. Not only are many of today’s government systems antiquated, they are also expensive to maintain and manage. The core systems have undergone so many changes over the years that the source code has become virtually obscure.

Governments worldwide are beginning to emphasize that their IT departments adopt Open Standards because of the improved interoperability, organizational flexibility and responsiveness that such an initiative can result in, and also as a means for avoiding vendor lock-in. As technology becomes increasingly an integral part of other disciplines, this new-found preference for Open Standards is driving innovation in politics, healthcare, disaster management and countless other sectors.

With its huge potential for saving money and improving operational efficiency, virtualization has come as a boon to cash-strapped Federal CTOs. Not only can it substantially reduce the cost of running data centers and corporate networks through more efficient use of both hardware and software, it can also significantly reduce the number of physical devices on the network, thus considerably lowering the complexity of managing the network infrastructure.

The Obama Administration has taken a hard look at the Federal IT infrastructure, and has come to the conclusion there’s way too much of flab that needs to be trimmed. Federal CIO Vivek Kundra has been put on the job, and he’s got his task cut out – reduce information technology costs, lower energy consumption, minimize IT real estate space utilization, bolster information security, and expand the use of cloud computing. Kundra has laid out a roadmap that proposes to:

In March 2010, exchange of patient medical information at VA-DOD had to be shutdown because errors kept popping up. In May 2008, heavy flooding forced evacuation of 176 patients from Mercy Medical Center in Cedar Rapids, Iowa as flood waters and sewage seeped into its basement. They barely managed to save their medical records. In May 2007, a massive storm ripped through Greensburg, Kansas and razed the Kiowa County Memorial Hospital, along with 95% of the town. The patients and staff were rescued. It was initially thought that all 17,000 patient records had been lost.

Recent times have seen frequent data center outages that have seriously affected the day-to-day lives of ordinary citizens. The latest to hog the headlines was the one at Virginia Information Technologies Agency’s (VITA) Storage Area Network (SAN) located at a large suburban Richmond Computing Center. The problem began on August 25, 2010 with the crash of a pair of three-year-old EMC DMX-3 memory cards. Technical glitches that grew from there soon mushroomed into a weeklong ordeal that affected 485 of the 4800 odd servers at the Center.

One very important aspect of successful project management is the creation of a Statement of Work (SOW). A Statement of Work can be defined as a narrative description of the products and services to be provided to a client under contract. Basically the SOW tells “what” needs to be accomplished rather than “how” it is to be accomplished, and clearly defines the scope of the project. Getting everyone to agree on the scope of a project at the very outset is important because it helps in minimizing scope creep.

There is a general misconception in the IT community that their database servers are somehow their most secure systems and that because they are secure their customers’ data and their own financial information is safe. Let me clue you in on this is a dirty little secret: many databases are neither secure nor safe from data tampering or theft. Although lots of data may be stolen by hackers for identity theft, a great deal of other data theft occurs from insiders who could benefit from sale of such data.

This article is written by Abhay Bakshi, who put together a book review of Ship it! A Practical Guide to Successful Software Projects.

I’ve noticed that most of my clients have finally been convinced that unit testing and regular automated acceptance testing is a “good thing.” However, I’ve noticed that load/stress testing, while moving up in importance, still has a long way to go. Even those doing load testing, though, forget to do proper network simulation to make sure that their tests will valid with “real world” network traffic.