This content is part of the Essential Guide: Prepare and manage enterprise apps for an IaaS model

Application workflow is key to cloud management success

To build a solid cloud management strategy, take a closer look at your application workflow and select the right mix of tools.

The purpose of managing an IT environment is to ensure an optimal level of cloud application performance and availability. In cloud computing, applications are separated from fixed resources, which means IT administrators in charge of cloud management have to manage the application workflow and resources, in parallel.

To streamline cloud management, start with your cloud application workflow. This allows you to determine application response time and identify the resources that support your applications. This can be done with familiar application workflow management and integration tools from major IT vendors like IBM, Hewlett Packard Enterprise and Microsoft.

Then, use the statistical tools available in your operating system (OS) and middleware to drill down into each cloud application resource. Augment these tools with data center management tools for your private cloud resources and with public cloud management tools for your public cloud resources. If your data center hosts a private cloud, consider a cloud management tool that handles both public and private clouds, as the market will continue to shift toward the hybrid model.

Since many organizations will shift to hybrid cloud eventually, look for hybrid cloud capabilities in any cloud management tools you assess.

When admins view an application workflow as an end-to-end process, it helps them determine response times for users, which is the starting point for cloud management. When they view it as a sequence of steps, the workflow helps admins identify the components that cause delays or interrupt processing. Cloud application workflow components are pieces of apps whose own statistics -- such as performance, status and packet loss -- as well as statistics for the associated OS and middleware, can help identify problems. Tools to access these statistics are largely available in virtual machine-based clouds, and there are also management interfaces from middleware, so you can start with those.

If you use multiple OSes and middleware platforms, your cloud management model will be more complex. However, there are tools available, such as NodePrime, to create consolidated views of a complex data center.

Considerations for choosing a cloud management tool

Even if you can get management data from your machine image OS and middleware, you need to supplement it with management data from your cloud provider. All public cloud providers have a management application programming interface (API) for basic deployment and health monitoring, and some, including Amazon, IBM, Microsoft and Salesforce, have a complete suite of cloud management tools.

High-functionality cloud management tools will generally address the technical and performance complexities that come with cloud. Since many organizations will shift to hybrid cloud eventually, look for hybrid cloud capabilities in any cloud management tools you assess.

Application performance management in the cloud age

Application performance management is more difficult and necessary than ever to the business. This webcast advises CIOs on how to get APM up to snuff.

Most cloud management tools help organizations build a single user and application interface on top of their specific OS, middleware, cloud platform and network management APIs. Beyond this basic model, the tools differ in focus. Scalr, for example, provides policy-based lifecycle management and automation capabilities for multicloud environments. RightScale is another tool that takes a top-down, role-based approach to what it calls Cloud Portfolio Management.

Cloud management road leads to DevOps

A streamlined cloud management model is a convenient jumping off point for DevOps. It can help you employ resource-specific DevOps tools and integrate them at the workflow level. If you want to run your applications in multiple environments -- such as the data center and across cloud platforms -- try not to use different DevOps tools for each of those environments, as it will complicate lifecycle management. Instead, consider a single DevOps tool that can integrate with your different IT environments.

A cloud management strategy is a top-down process that starts with applications, because they embody the business' goals. Think in layers. Commit the largest amount of money and staff to the layers closest to your business for an optimized cloud management strategy.

Next Steps

Compare eight cloud management tools

Demand for cloud-based management systems grows

Plan and manage a multicloud deployment

Dig Deeper on Cloud automation and orchestration