This the 7th part of our series on DevOps.
Check out Part 6: The Importance of DevOps in Enabling Compliance in AWS Environments.
Reliability, in the cloud technology era, can be defined as the likelihood that a system will provide uninterrupted fault-free services, as it is expected to, within the constraints of the environment in which it operates. According to the Information Technology Laboratories (ITL) bulletin published by the National Institute of Standards and Technology (NIST) on Cloud computing, reliability in Cloud environments is composed of the following:
- The hardware and software services offered by the cloud service provider.
- The provider’s and consumers’ personnel.
- The connectivity to the services.
In other words, reliability is dependent primarily on connectivity, availability of the services when needed, and the personnel managing and using the system. The focus of this article is about how DevOps can improve reliability when deploying in the Cloud, particularly in Amazon Web Services (AWS) environments.
DevOps can, in fact, help improve the reliability of a system or software platform within the AWS cloud environment. When it comes to connectivity, there is the Route 53 domain name system routing service, the Direct Connect service for dedicated connection, and ElasticIP, offered by AWS. For assurance of availability, AWS offers Multi Availability Zone configurations, Elastic Load Balancing, and Secure Backup storage solutions with Amazon S3. DevOps can be used to configure, deploy and manage these solutions and the personnel issues for reliability are practically non-existent because of the automation that is brought about by AWS DevOps offerings such as OpsWorks.
OpsWorks makes it possible to maintain the relationship between deployed resources persistently, meaning that an IP address that is elastically allocated to a resource (for example, an EC2 instance) is still maintained when it is brought back online, even after a period of inactivity. Not only does OpsWorks allow an organization to configure the instance itself, but it also permits the configuration of the software on-demand at any point in the lifecycle of the software itself, taking advantage of built-in and custom recipes. Deployment of the application and the infrastructure is seamless with OpsWorks as the applications and infrastructure can be configured and pointed to a repository from which the source is retrieved, automatically built based on the configuration templates, and deployed with minimal to no human interaction. Additionally, the auto-healing and automatic scaling features of AWS are perhaps what gives AWS deployments the greatest reliability, since when one instance fails, OpsWorks can automatically replace the failed instance with a new one and/or scale up or scale down as demand for the resource fluctuates. This ensures that there is uninterrupted and failure-free availability of services offered to connected consumers.
So, with a DevOps approach to deployment of cloud apps in an AWS environment, it may very well be possible that while the shift to Cloud computing may test the reliability of the technology, in reality, it is a not something that one needs to be overly concerned about. Try the DevOps approach to deployment and see for yourself how reliable it can be and is.