alex_aldo - Fotolia

Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Which quality assessment strategy is right for your cloud application?

Find the right monitoring and management tools for your cloud applications by choosing a strategy that works for your specific needs. Use the data to assess the overall quality of experience of your applications with application driven analysis

Which quality assessment strategy is right for your cloud application?

Unlike a dedicated server or virtual machine, relationships in the cloud between applications and their resources are highly variable. So application quality of experience (QoE) can vary, depending on the specific resources assigned and how those resources are connected. To that end, resource monitoring is difficult to do while also correlating with application performance.

Because of the complexities, it is important to understand monitoring and management tool options and how to pick the best strategy for your application. You need to know what it is that you're going to monitor and where you are going to do it. Point-of-experience (PoE) monitoring provides information on response time and QoE, but it doesn't offer any indication of where a problem might be occurring or how it might be fixed. Service border monitoring provides information on the behavior of a cloud service, based on the parameters that the cloud provider makes visible, but visibility into the cloud's resources is rarely available and so problem isolation and resolution is still difficult. Integrated application performance monitoring relies on gathering data from multiple sources and correlating it in a common operations center context to analyze resource state and application behavior.

Assessing QoE with application driven analysis

Hybrid cloud resource management starts with PoE data and adds the trio of cloud management data, networking management data and data obtained by deploying application monitoring tools as part of the cloud's application image. The first is obtained from client devices, the second from the cloud provider and the third is deployed as part of cloud application DevOps and integration.

Any abnormal condition from a management system is the start of a QoE exploration. From that source you move first to the "extremes," then back to the middle, meaning you look at the relationship between QoE at the point of experience and embedded application-level monitoring. This will likely identify the problem as being in the cloud or in the networking connection. It is important to note that hybrid QoE problems are most likely caused by poor resource distribution practices than by the failure of resources or network connections.

This was last published in July 2014

Dig Deeper on Network application performance

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.