IT Operations Management (ITOM)

Are your customer-serving VMs perfectly load balanced? Learn how to balance yours!

Are your customer-serving VMs perfectly load balanced? Learn how to balance yours!


Guest Post by Nagarajan K, HP Software Operations


In the era of mobile technology, organizations are excited about marketing their mobile applications. These mobile applications typically connect to the servers running in the organization's virtual datacenter. The success of these applications mainly depends on the performance, availability and scalability of the servers that are running in the datacenter.


Therefore it is critical that the related VMs running in the ESX servers are properly load balanced to make sure that, a sudden increase in the traffic/load does not disrupt the availability and performance of the server which is back ending the mobile applications.


There are great features, such as Distributed Resource Scheduling (DRS) which uses the vMotion technology to move VMs across the ESX servers belonging to the same cluster. What one should consider is that, the primary functionality of DRS is not to load balance the VMs across the ESX servers, but to make sure VMs from getting the resources which they are entitled to receive. DRS moves only those VMs which are deprived of the resources, even if there is an imbalanced usage of resources across ESX servers in the same DRS cluster.


When there is a sudden incoming traffic load, it needs some time for the DRS to identify such VMs and move it to another ESX server. This takes a considerable amount of time to perform, and the performance of the mobile applications suffers which is not desirable.


To protect the customers from performance and availability issues in case of a sudden spike in the load, you need a visualization and forecasting tool for the Virtual Infrastructure. You have to take special care of the critical servers, which provide services to the end users.


Thanks to Visualization and Forecasting tools like HP Virtualization Performance Viewer (vPV), which can help the administrators to visualize and redistribute the load on their Virtual Infrastructure manually.


You can download HP vPV to try it yourself here!


vPV has alerting mechanisms to detect an anomaly situation that can happen in a Virtual Environment such as abnormal (or deviation from its base-lined pattern) memory usage of a Virtual machine. The alerts can be a pro-active way of preventing “Out of Capacity” situation that can cause disruption of services hosted.


In virtualization, where resources are shared, vPV gives early indications on Resource saturation on Hypervisors or Disk resource.  From the usage pattern of a resource usage over a long period, vPV allows the admin to plan the capacity of their crunched resource using forecasting ability.


To learn more about this great tool, please visit the vPV Home Page.

  • infrastructure management
About the Author


This account is for guest bloggers. The blog post will identify the blogger.