Ten things project managers learn from off-the-shelf deployment tools

Off-the-shelf deployment tools have disadvantages, but even without buying in, project managers can learn a lot from the pros who build the tools.

There is a huge range of service providers anxious to help organizations succeed with infrastructure, platforms, applications or other systems deployments. It's not possible to tell just from reading the marketing hype which ones actually deliver on their promises. You need to dig through the information provided by reputable reviewers and satisfied (or dissatisfied) customers to draw your own conclusions. What we can glean from the features and benefits of these various products is a clearer picture of the complexities of deployment. This in itself can be beneficial in helping craft an application deployment strategy -- whether you decide to do it in-house or with the assistance of a third party.

Cost reduction and speed to production as end goals

All IT services firms have one thing in common. They claim to help businesses deploy on time and within budget, speeding time to value and increasing ROI. That's the Holy Grail of all IT projects and something that every business decision maker and stakeholder wants to hear. So, we'll take it as a given that these are your objectives as well.

For the purposes of this article, we'll forgo mentioning any brand names. We'll just focus on methodology, deployment tools and aspects of deployment that service firms target. We will look at what they claim to do and why you should make sure it gets done.

  1. Match business objectives and support business processes. Each deployment must make the business run better, decrease risks or increase profitability. If it doesn't, why bother?
  2. Reduce complexity and maximize operating efficiency. This may involve determining the optimal configuration for your individual environment or finding ways to reduce network traffic to ensure a snarl-free deployment. Diagnostics and assessment prior to the pilot will help you identify, isolate and resolve performance degradations and other issues. This will benefit IT after the fact by reducing service delivery problems.
  3. Ensure complete documentation and transfer best practices and other critical knowledge to IT staff. Everyone needs to have at least a baseline knowledge of what they are dealing with. This ties in to the three watchwords when handing off a project from one IT team to the next or to the end user: visibility, accountability, accuracy. These three features ensure errors cannot be easily swept under the rug.
  4. Integrate hardware, images, applications, peripherals and documents. Deployment isn't completed until all the moving parts work together. Detailed, accurate deployment diagrams help ensure that nothing is overlooked on the production environment side.
  5. Mitigate risk. This one is very high on the list of objectives that are important to business stakeholders. Your deployment should ensure that downtime is very limited, service levels are high and there is minimal disruption to business activities during installation. Ideally, the technology or program that you deploy should increase availability of applications.
  6. Use proven methodologies that entail a high level of automation. Any deployment expert will tell you: Manual may still be OK for high-end sports cars, but it won't give you a high-performance deployment in the technology sphere. However, automating without knowing what you are doing means you will have to go back and fix stuff sooner than if you did everything manually. It's not exactly "failing early," just "failing often."  Proven automation tools and systems such as those used to package applications can and do cut down on technician intervention.
  7. Minimize network bandwidth for large deployments via factory pre-staging and process automation. This becomes an issue when choosing a manufacturer for your laptops and desktops. Some of the larger equipment providers offer these services to ensure a swift, trouble-free deployment experience once the equipment hits your dock. They may include image configuration, application configuration, data migration and settings migration. If you are doing a software deployment that is predicated on the correct setup of new equipment companywide, you can see how useful that would be.
  8. Transfer and manage critical data. In the rush to deploy an application, it's all too easy to forget that the application is useless unless it can access the right data in a timely fashion. With a migration from legacy to a cloud-based infrastructure, maintaining the integrity of your data is even more difficult. Any deployment plan must take into consideration how the deployed technology or software interacts with or impacts your databases.
  9. Boost user adoption. Full-service IT firms focus on making sure your deployment is met with eager participation instead of grudging acceptance by end users. Intuitive interfaces and a gradual introduction to new features can really help here.
  10. Manage de-installation and software or equipment disposal. Every deployment of something new involves saying goodbye to something old. Or, at least it should. Keeping useless files/directories or outdated equipment leads to a cluttered virtual and physical environment and makes maintenance more difficult. Any deployment strategy should include an "Out with the old, in with the new" plan.

If you feel that your current resources aren't up to accomplishing these 10 things, it's probably time to call in the experts!

Dig Deeper on Java application deployment

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.