Network and IT infrastructure management is moving towards achieving data center agility: agility in deploying new application services and agility in troubleshooting network problems that cause application downtime. True agility in the infrastructure can be achieved only when teams have better visibility (state, status, and statistics) of the infrastructure and change control has certain level of automation enabled.
Application-centric infrastructure helps network engineering and operations teams respond quickly to application downtime, yet automation is the key to ensuring quick, error-free changes to the application infrastructure.
Change automation ensures better predictability and availability of applications as changes are planned, designed, and implemented during change windows. The IT Process Institute’s Visible Ops Handbook reports that “80% of unplanned outages are due to ill-planned changes made by administrators or developers.” A recent Gartner study projected that “Through 2015, 80% of outages impacting mission-critical services will be caused by people and process issues, and more than 50% of those outages will be caused by change/configuration/release integration and hand-off issues.” The lack of process and visibility in change management leads to large number of misconfigurations in the network and ever-declining customer satisfaction. Automation of change management ensures better processes through approval workflows and integration with enterprise ITSM tools and change automation platforms.
So, what is necessary to achieve this so-called “automation”?
1. A centralized solution for change management: All changes need to go through one window to allow easy tracking of all the changes done across the infrastructure. Audit trail reports of changes enable easy detection of configuration drifts in the network and easier troubleshooting during application downtime.
2. Customizable workflows: Changes need to go through multiple levels of review before the implementation team pushes them to the devices. A production environment may require more approvals for a change compared to development or pre-production environments. The workflows need to be customizable to ensure each request is handled with the right kind of focus. The more reviews there are, the fewer misconfigurations and the higher the application availability there will be.
3. Integration with enterprise change systems: Changes outside the allocated windows are the biggest cause of application downtime. Integration allows for changes to be planned to ensure better application predictability.
4. Application-centric configuration management: Application teams should be able to map any configuration drifts in the application infrastructure. Which firewall policies got changed? Which configurations were modified in the load balancer? All changes need to be tied to an application and the application operations teams need visibility into the service infrastructure.