Blog

3 Essential Areas for Performance Monitoring in VMware’s Hybrid Cloud

June 12, 2016 | Heroix Staff

Moving some or all of your infrastructure to VMware’s vCloud Public Cloud will add in a couple of new facets to your performance monitoring, but won’t remove the need for the monitoring tools you’ve been using in a physical environment.  In a VMware Hybrid Cloud environment - using VMware in both the public and private cloud - performance monitoring can be broken down into 3 essential areas:

1) Server Based Application and Performance Monitoring

Virtual servers don’t know they’re not physical servers.  That’s high up on the list of appealing features for many system administrators:  you can build servers and configure applications pretty much the same way that you would for a physical server, but you’ve got a lot more flexibility about how to assign and change the resources for the servers.  But, servers still need to be monitored – resources are not dynamically added to a server if you haven’t allocated enough memory or CPU to start with, or if the server outgrows the resources you’ve given to it.  You still need to monitor performance metrics and system logs.
The same goes for Application Monitoring.  Just because the server hosting an application is running and not spiking resources doesn’t mean that the application itself is performing well.  SQL can still have blocked processes, web pages can be hacked, Exchange mailboxes can fill up….cloud based applications suffer from the same ills as applications on physical servers, and need to be monitored accordingly.

2) Network Monitoring

Your WAN is standing between you and your public cloud, and anything that slows down your WAN will slow down your application performance.   In addition to monitoring bandwidth, latency and uptime between your LAN and your public cloud servers, you should also consider SNMP Trap based alerts for Network Devices, and, if enabled, forwarding Syslog records from the devices to a monitoring device.  Both SNMP Traps and Syslogs can provide important early warnings for network device hardware problems that might otherwise only be found through a detailed analysis of performance metrics, or when you start receiving calls from frustrated users.

3) VMware Monitoring

If you’re new to VMware, this may be a new set of tools for you.  Monitoring VMWare is done on two levels – first, to check that all the VMware components are functioning within specs, and second to build a baseline to use for capacity planning.  When you initially provision your VMware Virtual Machines (VMs), they’re provisioned with your best estimate of the resources they will need.  Over time, with the creation of additional servers, and with some servers having too many or too few resources, you are likely to find VMware sprawl creeping in – where your carefully planned out initial VMware design has been bloated by too many VMs and over-provisioned VMs.  A VMware capacity planning baseline will help you determine how to re-allocate resources, and which VMs are no longer needed.

In order to maximize the increased uptime and scalability of Cloud based servers, you need a performance monitoring solution that will cover the essentials for not just server and application performance monitoring, but will also factor in network and VMware monitoring.   In the cloud, monitoring is no longer just about server or application performance – you will need to look at the environment as a whole.

Want to learn more?

Download our Virtualization or Cloud IaaS Whitepaper - both technologies can provide redundancies that will maximize your uptime and that will allow you to squeeze out the most performance. Which is better and how do you decide?

 

Download the whitepaper:  Virtualization or Cloud IaaS?