With traditional data centers, IT teams spent the bulk of their time on manual system provisioning and management. When a new server arrived or required an upgrade, it was a complex, time-consuming process. Cloud computing, however, changed this scenario significantly.
Today, system provisioning has been reduced to a few keystrokes. But, as cloud adoption has risen, additional provisioning requirements have emerged, such as integration among multiple platforms, which often means new challenges for IT teams.
The evolution of the cloud provisioning process
Automation was the first step to accelerate and improve the provisioning process -- both in the cloud and on premises. Through software, IT teams complete a particular task, such as launching a web server, with minimal manual effort.
But as IT deployments grew in complexity, organizations yearned to do more than automate one function in a piecemeal fashion; they wanted to streamline their provisioning workflow, which required the automation of multiple tasks across multiple systems.
Orchestration tools, such as Ansible, Chef, Puppet and Salt, emerged to meet that need. The goal of these tools is not to just automate one function, but to deliver a service. For instance, Puppet Enterprise automatically determines the correct order of operations, what information needs to pass securely between services and when to wait for a service to become available to fulfill the next step in the cloud provisioning process.
Cloud applications were built to take advantage of such features, including automation and orchestration. Private cloud services, specifically, have also become easier to provision as vendors like VMware and Microsoft improve their management tools. VMware's vRealize Orchestrator, for example, now includes a library of prebuilt scripts that speed up the creation of new cloud workloads.
Public cloud platforms also offer high levels of automation and orchestration, and vendors constantly improve their functionality. Amazon Web Services, for example, updates its services continuously, and its CloudFormation tool includes templates that organizations can use to orchestrate resource deployment.
New obstacles, new cloud provisioning tools
As a result of these advancements around provisioning, automation and orchestration, enterprises can roll out cloud deployments more quickly. However, challenges remain.
For instance, various cloud management tools typically do not meet all enterprise requirements out of the box. To deploy the tool in a way that meets their needs, IT shops often need to customize them and work with APIs to invoke higher levels of orchestration.
Management issues also arise, especially as organizations adopt more than one cloud service. IT wants to use one central management console, but that's not always realistic. Even if private and public cloud provisioning processes are automated, integration between them can be a challenge.
A number of third-party cloud provisioning tools have emerged to help overcome these challenges. Some examples include:
- CloudBolt: A tool designed for hybrid environments that can consolidate internal and external clouds into a single, user-facing application. Users can then manage and provision resources on demand, and administrators can set provisioning conditions that limit user access to only the resources needed for their jobs.
- Embotics vCommander: A self-service provisioning and management system that enables users to manage VM and application workloads. Through a web-based service portal, users request new workloads or changes to existing workloads.
- Morpheus Data: A tool that enables companies to allocate databases, apps and app stack components. It automatically collects system, database and application logs for all provisioned cloud systems and provides reports to notify admins of any performance or uptime issues.
- IBM Cloud Orchestrator: Based on OpenStack software, this IBM tool works with private clouds that run VMware, Hyper-V, Kernel-based Virtual Machine, Power and z/VM Infrastructure, as well as public cloud platforms including Amazon Elastic Compute Cloud, IBM SoftLayer and Microsoft Azure. Its dashboard provides users with an overview of cloud application health, and the tool offers what-if analysis features related to capacity.
- RightScale: A tool that offers template-based provisioning capabilities for private, public, hybrid, bare-metal and virtual clouds. A catalog of templates is designed to speed up the cloud provisioning process and ensure policy consistency.
While these tools offer potential benefits, they also have limitations. For example, they're designed primarily for large, complex environments, according to Dennis Smith, research vice president at Gartner. This means they still require a fair amount of integration work and can be expensive, which makes them out of reach for some small and medium-sized businesses.
Create a simple self-service provisioning portal
Seven cloud management tools admins need
Avoid overprovisioning with right-size techniques