By 2016, 41 percent of all enterprise communications application users worldwide will have migrated to the cloud according to a study by ABI Research. That translates into trillions of dollars in business revenue depending on the delivery of these services. Application performance management will become even more critical to daily operations, but surprisingly only a small number of cloud-based application users have adequate performance management software monitoring their service delivery today. Why is this a problem?

Cloud computing is a moving freight train, but not with all the parts moving in the same direction at the same time. This is typical of any emerging technology or services. Today, many IT organizations within enterprises have stated plans to migrate to the Cloud to improve operational costs efficiencies and to provide better services to the business users. There are also a number of IT pilot projects utilizing Cloud services. The most pressing issues to organizations today – and the ones holding back increased investment – are security, compliance and performance, in that order. As Cloud technology matures, there is no doubt security and data privacy issues will and are being resolved. However, Cloud application performance is a different beast and something equally as important to ensure adoption and success.    

In today’s non-Cloud environments, application performance monitoring is going mainstream, according to the Gartner Hype Cycle for IT Operations Management, 2011. Gartner had seen an increase in demand from their clients who are transforming from purely infrastructure management to application management. In today’s non-Cloud environment, with the current technology for application performance management, it is possible to instrument and collect run-time metrics and provide access to management tools to analyze and report on the metrics. Thus, it is possible to obtain a comprehensive view of an application including end user experience, specific transactions, and the supporting delivery infrastructure in order to manage the availability and performance of a business service.

When parts or all of an application moves to a Cloud, the view into the application is disrupted. This may be due to the loss of access to instruments for metrics collection, or access for the data to make its way back to the management system. But, here lies the problem with performance management in the Cloud – there has historically been a disconnect. One thing that doesn’t change for both Cloud and non-Cloud environments is that the users, representing the business, expect the same level of availability, access to the applications, and performance. This presents a new challenge for organizations looking to leverage the Cloud.

At a high level, the Cloud infrastructure includes the application delivery infrastructure that is made up of the applications running in virtualized environments and the network that supports the delivery of the applications to users. While the infrastructure itself is still made up of switches, routers, firewalls WAN optimization devices, VM Hosts and servers, the new element is that there could be more than one owner, such as cloud service providers and the private enterprise, for different parts of the delivery infrastructure.

Let’s look at these owners and how they can address application performance management in more detail:

Private Enterprise - For an enterprise IT organization that is building a private cloud and virtualizing applications, there are some key challenges to evaluate when looking for an application performance management solution:

1.      Bridge the Gap– Since application and network performance management bridge between cloud and non-cloud environments, it’s likely an organization is not moving all applications to a private cloud.  Whether in transition to migrate applications or simply maintaining both cloud and non-cloud based applications, the challenge of managing availability and performance for both sets of applications exists. It is crucial that cross-functional IT team members have a common, unified system that they can rely on for problem domain isolation, root cause identification with actionable information so that they can resolve the problems.

2.      See More– Within a private cloud, the challenge is visibility of applications in a virtualized environment. To accomplish this, an organization needs flexible data collection instrumentation. It is important that the instrumentation can measure the performance of multi-tier applications as well as provide transaction level information for root cause analysis when performance degrades. This requires supporting deployment models to see the intra virtual machine’s traffic. The graphic below is an example of where Visual Network Systems’ OmniPoint (from Fluke Networks) resides to provide this visibility.

3.      Be Scalable and Prepare for the Future– As with any new technology, it’s important to collect new information. The chosen solution needs to be extensible to support new relevant performance metrics without having to do a mass rip and replace. A proven scalable architecture is important especially if managing many remote offices. For the IT team to be effective, the architecture needs to be able to support mediating a variety of data sources in a delivery infrastructure and correlating performance metrics to provide a comprehensive view of application performance.

Service Provider - For the public cloud service provider that is anticipating the demand from enterprises to prove performance of application delivery with Service Level Agreements, it’s important to consider the following when leveraging a performance management solution:

1)      Be Flexible– The instrumentation should offer complete visibility of cloud infrastructure and support problem domain isolation (enterprise, carrier or cloud provider). Since cloud infrastructure is a dynamic environment, it means the instrumentation should be deployable anywhere and supports a host of platforms. This also includes deployment to get visibility of intra virtual machine traffic and VM resources usage. The yellow dots in the following graphich demonstrate possible instrumentation locations to ensure this visibility.

2)      See All and be Ready for Change– Service providers are supporting hundreds if not thousands of customers with their infrastructure. It is imperative that the performance management solution chosen has proven scalability in a service provider class network. This means the architecture has to be able to handle collecting performance metrics from many instrumentation points and correlating the data to present relevant views of the information. Since it is undesirable to have large amounts of unsolicited data traversing your infrastructure, the architecture has to support minimizing traffic back to the management system, but still provide a complete view of the performance of the infrastructure. As with any new service or technology, the infrastructure will be changing. The solution needs to be built on a platform that is easily extensible to include future new measurements without costly replacements.  

3)      Report for All Stakeholders– Enterprises will be demanding Service Level Agreements from their service providers and a view of the performance of the services they subscribe. The solution needs to be capable of collecting and reporting service level metrics, such as end user response time, traffic usage, and volume for each enterprise accounts. The architecture of the solution needs to be multi-tenant, meaning, it can collect, analyze, host and report on performance metrics for each individual enterprise account.  

As more and more organizations leverage the cloud to support and extend infrastructure, managing application performance will play an increasingly vital role for success. Whether using a hybrid private cloud, or cloud service from a provider, it’s key that organizations have insight into whether or not application services are being delivered according to availability and performance objectives.

As with any new technology, there are pitfalls if expectations are not set properly and the transition not managed carefully.  As these services mature, and cloud adoption becomes mainstream, so will the understanding of what it means to truly deliver quality end user experiences in the new environment. Application performance management is the key to ensuring an organization can track, manage and resolve these issues before they impact the end user and the organization.

ShareThis

Tags: Hosting & Colocation

Read more White Papers»

More DCS TV»

More Audio»

Read more Web Exclusives»

Read more Magazine Articles»

Read more Supplements »

Keep up to date with the latest industry products, services and technologies from the world's leading IT companies.

• Virtualization Success Depends on Network Automation by Infoblox

• The Infoblox VMware vCenter Orchestrator Plug-in by Infoblox

• Automated Storage Tiering on Infortrend’s ESVA Solution by Infortrend

• Why Cloud Backup: Top 10 Reasons by Autonomy

• How to Measure the ROI of Cloud Data Protection by Autonomy

Latest IT jobs from leading companies.

Click here for full listings»