Experts largely agree that automation is an important ingredient for managing cloud services and technology. Typically, this means a battery of pre-defined processes and procedures run according to a well-defined schedule and set of policies. It also means adopting a set of best practices to make sure that you stay on top of resource consumption, errors, alarms and alerts, and compliance matters.
Cloud Technology Metrics
In planning, designing, implementing – and, of course, managing – cloud technology, metrics are not only important, they can spell the difference between success and failure at many levels. Certain items, including the following, need to be baked into the cloud lifecycle process from initial design to its end:
- Bandwidth consumption*
- Application/webpage latency and response time, usage and visitation tracking
- Uptime and availability percentages
- Cloud resource consumption: CPU, storage, infrastructure (SDN, NFV and virtualization)*
Asterisked items usually provide the basis for pay-as-you-go charges from cloud service providers in the public cloud, and to application owners or responsible organizational units for chargebacks in private clouds.
Service-level agreements with public cloud service providers may focus on other items and will often be tracked in dashboards or management consoles to make sure buyers of cloud services get what they pay for. For this kind of thing, automation is key to generating statistical data for tracking and analysis, and compiling reports so stakeholders, managers and executives can keep up with consumption, usage and related costs.
Automating Cloud Technology Essentials
Automation for the cloud really comes into its own, however, in several areas of activity. First and foremost, automation is essential to handle the setup and tear-down of virtualized applications, platforms, services and infrastructures.
In fact, there’s usually a huge amount of detail work involved in configuring, provisioning and deploying added capacity or capability when additional cloud assets or resources must be laid on, during peak demand cycles or to support demand-driven growth. The same is true when demand falls off, and virtualized resources must be pruned to avoid excess costs for idle capacity or capability.
The volume of activity involved means it is neither possible nor practical to assign such work to humans. They’re simply not fast enough, and they’re too expensive to handle such things in real time as workloads wax and wane. This kind of automation is really what makes cloud technology such an attractive proposition for businesses and organization, especially at large scales.
Proper automation lets IT workers create rules or specifications for cloud technology that drive how cloud solutions proliferate and situate themselves within the cloud and lets the tools generate the scripts or programs that do the actual work involved.
This covers scenarios that include fluctuations in demand, adding new or removing existing cloud services, components or applications, and responding to cyclical or seasonal market conditions. In a very real sense, this kind of automation makes cloud technology possible.
Programmed Responses in the Cloud
Another important area for automation falls under the heading of programmed responses to various triggers.
These include a broad range of events, such as the following, that might require millisecond-level reactions:
- Changes in network link status
- Users leaving or entering the cloud
- Errors or failures in the network, servers, applications, or storage
- Shift changes causing moves from call or data centers going offline to those coming online (following the sun around the globe is an easy way to understand this)
This kind of triggered automation also applies to the following types of events:
- Detecting and dealing with security threats
- Handling failover situations like those associated with business continuity and disaster recovery scenarios
- Moving workloads from the private cloud to the public one, or vice-versa, for business or cost reasons
This is what keeps the cloud workable and cost effective over time – especially combinations of one or more private clouds, and one or more public clouds (called a hybrid multicloud).
Getting on to the Cloud, or Moving Among Clouds
These days, most companies and organizations are already invested to some degree in cloud technology. Looking forward, however, the tendency is to move more and more computing and data into public and private clouds.
Automation helps handle the huge volume details involved in picking up and moving applications and services from on-premises implementations into public and/or private clouds. Ditto for migrating workloads (and their supporting APIs, middleware and plumbing) from one cloud to another. This usually involves plenty of real programming work as well, including open technologies designed to let companies and organizations avoid vendor lock-in to a specific cloud service provider’s APIs. For example, Red Hat’s OpenShift Container Platform works across Amazon Web Services (AWS), Microsoft Azure and Google Cloud Platform to support cloud-agnostic application integration. Related automation tools provide management data and insight through key metrics and visualizations.
Cloud Technology Jobs: What Do IT Pros Need to Know?
To stake out career opportunities for themselves in this rapidly growing and high-demand technology area, IT pros should equip themselves with a good general understanding of automation for DevOps and cloud management. They should acquaint themselves with open technologies for containers and cloud abstraction, and one or more of the big public cloud provider’s platforms (AWS, Azure, Google, etc.).
Then, they should learn two or more key automation languages, with one being Python because of its extreme compactness and expressive power. Opensource.com, lists five of these languages for DevOps and cloud management:
- Python
- Ruby
- JavaScript
- Go
- C
Chef and Puppet are also included in copious job postings.
Gaining these skills will prepare IT pros to roll up their sleeves and help a current or prospective employer use automation to make the most of their cloud investments.
CompTIA Cloud+ validates the skills needed to work with cloud technologies. Download the exam objectives to see what's covered.
