integration and implementation of technology-focused business solutions

September 2018

When something goes wrong in IT

September 18th, 2018

It’s important to talk about the risk of a project up front. Many clients we meet with don’t like that part of the conversation. They think nothing could ever go wrong, but that’s not the reality of the situation.

Any number of things can change the scope of a project unexpectedly. In some cases, a part the customer thought was done by another vendor may be way more complicated than anyone ever imagined. Or even worse, it could be completely missing.

There are times when something doesn’t work like anyone thought it did work or should work, and you basically have to pull out all the wires and rewire it.

So how do you handle it when something does go wrong? When you run into that situation you talked about at the beginning with a client?

Not long ago, we ran into a situation like that. We picked up a project originally developed by another vendor. Once we got in there and started the work, we discovered it was a mess.

Some of the people we were working with in the organization weren’t surprised. Not because of the original vendor specifically, but because it’s part of the nature of any IT project.

However, when we sat down to talk with the client, they let us know we still couldn’t go over the budgeted amount.

Of course, we were willing to work within that. But when there are budget constraints, the scope of the project has to change. We weren’t approaching it trying to sell them on extending our services. We just needed to have a conversation about how to finish the project when that unexpected thing comes up.

When the unexpected happens, be open to talking through possible solutions. You may have to increase the budget to get what you originally wanted, but you may also be able to eliminate or delay certain features or functionality.

The point is, you have to make adjustments and have open communication. Talk through what moving forward on the project is actually going to look like. Because if you don’t have the conversation when the issue is first uncovered, it will be even more of a mess once you get to the end of the project or budget.

Comments: None

Continual rollout of self-driving databases from Oracle

September 4th, 2018

Oracle continues to roll out features to its cloud-only product, Oracle Autonomous Database Cloud. Their patch process will roll out features on an ongoing basis to ensure the database is on self-driving mode.

Let’s look at some high points about these self-driving databases.


You’re going to get new features, improvements, and fixes much faster in the database world than you ever would in a car. Oracle has already announced that security patches will automatically be applied each quarter, which is much faster than most manually operated Oracle databases.

And when it comes time to upgrade or patch, the Autonomous Database can apply the real production workload on a test database to be sure there are no unexpected side effects.


Traditionally, Oracle has not been known to give guarantees. But with this product, they’re actually giving two of them.

First, they’re guaranteeing downtime is limited to 30 minutes a year, including maintenance. Second, they’re guaranteeing that they'll beat Amazon's price for AWS by 50%.

Need for DBA

Of course, just like a self-driving car still needs a driver, the Oracle Autonomous Database Cloud still needs a database administrator. However, the overall hours needed to perform routine DBA tasks will be greatly reduced.

Let’s say, for example, that you had ten database systems running in the cloud already. You would probably no longer need two database administrators for that, so one of those individuals could take on a new role or additional duties.

And if you have just one database administrator, this will free up their time to work on other projects. I can’t tell you how many times over the years I couldn’t work on a project because a third of my time was taken up with basic database administration tasks.

Routine tasks around performance, storage, and uptime will continue to be automated, which will eliminate many of those from a database administrator’s to-do list.

Comments: None


501 E. 15th St., Suite 200B
Edmond, OK 73013