Deployment
Deploying a change to a platform that is currently running on 400 production servers.Monitoring
Monitoring user experience such as response times across a cloud platform where users are connecting to different machines in different data centers.Incident Management
Identifying a production problem and correlating it to a server, set of servers or a particular type of configuration.Configuration Management
Keeping an accurate history of configurations across complex environments.Rollback
Reverting a large number of servers back to a historical configuration.Security
Applying a security fix to a device driver on thousands of servers running different systems and applications sponsored by different business units.Administration
Automating cumbersome administrative tasks such as daily backups across a large number of devices.Alerts
Generating and handling alerts. For example, a large telecom company might receive thousands of alerts every few minutes and they need to automatically figure out which indicate a problem that actually impacts customers or end-users.Provisioning
A data center needs to quickly change the linux version on 50 physical machines to meet customer demand.Capacity
An internet service needs to balance load across 7 data centers.Resilience
A cloud service has an average of 7 servers fail in various ways each day. This needs to be immediately detected and handled by redirecting load to working servers.Disaster Recovery
A cold backup site needs to be continuously synced with production.Overview: Infrastructure As Code | ||
Type | ||
Definition | The automation of IT infrastructure deployment, monitoring and management. | |
Notes | In the past, infrastructure teams were mostly specialists in hardware and operating systems who focused on connecting and configuring things. Infrastructure as code represents a shift towards having infrastructure managed by developers. | |
Related Concepts |