Public and private cloud computing have received plenty of attention in recent years, as businesses worldwide have opted to implement such services. For small firms, the public model is a boon for cash-strapped companies that cannot afford any missteps in their IT procurement processes.
Large enterprises that want more control over their solutions have often opted for a private cloud, which is not a shared resource with other businesses. Although both of these options are ideal for their respective user base, hybrid clouds should not be overlooked in the grand scheme of what is becoming a necessary technology throughout the IT sector.
A recent Infonetics Research survey indicated that adoption of hybrid clouds among enterprises is projected to more than double by 2015. Platform-as-a-Service and Cloud-as-a-Service are expected to experience the largest increases between 2013 and 2015. Cliff Grossner, directing analyst for data center and cloud services at the firm, said hybrid solutions are "the next evolution in cloud architecture."
However, the cloud abstracts important detail from the people who need to make sure it's performing as expected. As business-critical applications move to the cloud, IT professionals need to understand what's happening in the "black box" beyond their physical reach.
According to Enterprise Management Associates, only five percent of companies can definitively pinpoint the source(s) of their application-related problems – and these percentages relate to on-premises applications only. At the same time, more than 50 percent of companies report the cost of an hour of downtime for the “most critical business applications” to be between $75,000 and $500,000.
The goal is to give admins the ability to identify poor user experience before it becomes a costly business issue. Yet root cause analysis requires visibility into the underlying components of the application, which is hard to achieve when the infrastructure is owned by a service provider, or distributed across disparate monitoring silos. Organizations that are far down a path with on premises or private cloud APM solutions often don’t have the tools required to deliver real-time, proactive information in public or hybrid cloud environments.
Hybrid cloud APM addresses this problem, providing a single pane of glass from which to manage application performance and availability across public and private environments, from server to website to end user.
As more hybrid cloud application deployments go mainstream, end users must be able to expect the same level of availability, access to applications and performance in the cloud that they receive from non-hosted applications.
Here are some key features to look for in a hybrid cloud APM solution:
- Ability to collect and send performance metrics at any frequency
- Ability to create alerts on custom metrics to know the moment performance exceeds expected bounds
- Ability to customize dashboards for integrated views combining any metrics
- Ability to monitor multiple application services, such as Apache, MySQL, Redis, Cloudwatch, as well as create your own
- Ability to see data across all boundaries, including cloud or on-premises, various clouds, and separate regions/availability zones
Cloud computing is all about running the right application workload on the right system at the right time. As increasing numbers of workloads become split between in-house and hosted environments, organizations will need to carefully consider how their APM solution addresses cloud-related challenges beyond their immediate control.
Eric Anderson is CTO and Co-Founder of CopperEgg.