momius - Fotolia
As private cloud adoption increases, networks must run smoothly for enterprises to capitalize on private cloud's benefits. A few tricks to optimal private cloud network performance include choosing the right tools to keep performance in check and properly designing workloads for the private cloud.
Network and workload performance monitoring
A well-designed private cloud network doesn't guarantee optimum performance. Cloud admins must also monitor workloads to ensure their private cloud runs efficiently. Some tools can keep an eye on both the network and applications.
Network and cloud-based application performance monitoring (APM) tools work hand-in-hand with existing monitoring and management data center services. These tools can track performance, report results to IT teams and alert stakeholders to possible service disruptions.
Tools that automate workload migration are valuable to balance workload distribution across private cloud systems. Hybrid clouds allow some workloads to burst from your private cloud to a public cloud provider as computing demands and available resources dictate.
APM tools maintain performance levels by achieving a safe range of resource use to preserve the quality of service (QoS). IT teams can dictate private cloud QoS benchmarks, or they could be spelled out in service-level agreements that formalize a business relationship between IT and private cloud users.
Cloud admins can perform network monitoring through any number of locally installed system management tools, but the clear trend is to use third-party testing services for cloud APM. Popular tools include Soasta CloudTest, Apica Global Testing and Monitoring Network, testing and monitoring services from Keynote Systems, BlazeMeter cloud testing services and many others.
Design workloads for the cloud
An application's performance is inexorably connected to its underlying design. To achieve the best performance, apps need to be designed specifically for the cloud. Cloud applications should feature scalability to use or release resources according to demand. The cloud app also needs to be tightly coded and not easily affected by storage and network latencies.
Unfortunately, these characteristics are often different from those of traditional, locally installed applications, which often don't perform well -- if at all -- in a cloud. Many applications see substantial benefits from optimizations, or even a fundamental redesign, that supports native cloud environments and APIs.
App redesigns can be expensive and time-consuming. And they frequently are attempted with new software design paradigms, such as DevOps. But application design efficiency affects long-term cloud performance, especially as more applications are moved the cloud.
Stephen J. Bigelow is the senior technology editor of the Data Center and Virtualization Media Group. He can be reached at firstname.lastname@example.org.
All AWS eyes on Google after cloud monitoring buy
Application monitoring tools for Google cloud
How to endure a private cloud failure