Why has performance management become such a key issue in cloud computing? Well it’s simple: there is little point in moving to a cloud solution if end users can no longer use their business-critical applications effectively. We spoke to Orange Business Services expert Jean Critcher to find out about the four key issues you need to address when looking at application performance in the cloud.
build application performance into the business case
An assessment of application performance needs to happen at the very outset when you are building a business case for cloud computing. Application performance is a key factor in helping you choose what applications to move to the cloud. For example, if your main criteria for cloud adoption is costs savings, then any performance issues that slow your business processes or damage productivity could virtually eliminate any savings made.
determine how to judge success
To be able to effectively judge application performance you need to identify your key performance indicators (KPIs). These will also need to reflect the impact of different deployment approaches, such as public or private cloud. As a guide, a survey carried out by Orange in application performance KPIs found that application response time is the most common (93%), followed by availability (72%), transaction response time (68%), latency, jitter, packet loss (68%) and server response time (66%).
Monitoring KPIs should be an ongoing activity – even once the application has been deployed in the cloud – because they give you valuable information on how your applications are reacting to changing business and technical conditions, such as the impact on availability caused by a wider rollout, for example.
set your service level agreements
Your KPIs should also play a part in forming the service level agreements (SLAs) that you have with your service provider. Our survey found that application-oriented SLAs came in first place (69%), followed by managed network services (56%) and business service or systems management SLAs (27%).
The approach your service provider is taking to application deployment can have a big impact on how your applications will perform. Even the way the application is virtualized can make a big difference – by virtue of how customers are partitioned in the shared infrastructure. Because resources are not unlimited, the actions of another customer could have an impact on your own application’s performance. This should be reflected in your SLAs.
don’t do it all at once
Application performance in the cloud can suffer if you try to do too much at once. You need to take a phased approach to cloud computing, so that you can see how well your cloud application performs and work out the implications to your contingency plans and what it means to your business continuity planning (BCP). Once your application has been tested and is stabilized in a live environment, then you can look to add new applications to the cloud. Governance is a key part of this exercise. It gives you a framework for how applications are distributed, replaced and upgraded to prevent any changes having an impact on performance and availability.
Read more about the cloud in our recently published Cloud Computing Insight.
After a Masters in Computer Science, I decided that I preferred writing about IT rather than programming. My 20-year writing career has taken me to Hong Kong and London where I've edited and written for IT, business and electronics publications. In 2002 I co-founded Futurity Media with Stewart Baines where I continue to write about a range of topics such as unified communications, cloud computing and enterprise applications.