In my previous post, I focused on the APM challenges when implementing and running a VMware ESX server environment. In this post, I’ll discuss how to solve the APM challenges in a VMware environment.
As I covered in the first post, when applications are virtualized and collapsed inside a single piece of hardware, the “one-to-one” relationship between applications and hardware becomes a “many-to-one” relationship, and legacy monitoring solutions lose their analytical capabilities. Key challenges include:
- limited visibility into transactions, especially between VMs on the same ESX host
- limited visibility into the physical-to-virtual relationship between hardware and applications
- difficulty understanding the performance impact of Virtual Machine Managers (VMM).
Each of these has a serious impact on any APM solution that isn’t configured to operate in a VMware environment, so let’s take a look at how to solve the APM challenges in a VMware environment.
Application and Transaction Visibility: To provide complete visibility into application transactions, including those that are residing on the same ESX server, and to provide information on dynamic mapping of applications across ESX servers, there are two instrumentation choices:
1) Virtual Network Probe
A network monitoring application (the Virtual Probe Appliance) is installed on a VMware guest image. The virtual probe collects and communicates virtual network data inside the ESX server in exactly the same way as a physical probe collects and communicates physical network data in the real world. Virtual NICs (vNICs) on the virtual network are configured in “promiscuous mode” making the traffic visible to the internal probe.
2) Cisco Nexus 1000V Switch
An additional approach which is effective in larger, enterprise-caliber environments is to incorporate the Cisco Nexus 1000V Series Switch2 into your VMware platform. These are intelligent virtual (software) machine access switches that operate inside the VMware ESX Hypervisor and offer exactly the same set of features as Cisco’s physical switches — providing a single, common set of provisioning, configuration and networking features across the entire networking environment, both virtual and physical.
From an APM point-of-view, a key feature of the 1000V Switch is it enables port spanning, which means it can be used as the equivalent of a hardware switch. Attach an external network monitoring device, and you can analyze all virtual network traffic as if it were coming from a physical mirrored port.
Retrieving data from Virtual Machine Managers: Each ESX server uses a VMM (or “Hypervisor”) to manage the virtual machines running on that server. To facilitate the management of multiple Hypervisors across the network, VMware’s vCenter Server acts as a single point of information for virtual infrastructure administrators. It offers an extensive set of application programming interfaces (APIs), allowing third-party applications access to data collected by vCenter.
In my final post of this series, I’ll offerr a comprehensive APM strategy for VMware environments.