Garner’s Analyst Jonah Kowall recently published a research note discussing the future of End-User Experience Monitoring in APM (See End-User Experience Monitoring in APM: Past, Present and Future Published: 11 January 2012. Analyst: Jonah Kowall).
I thought that I should address the tactical issues of End-User Experience Monitoring in virtual systems. A big problem that was only mentioned briefly. I will.
There's one big impact to Kowall’s analysis: Convergence. Convergence of APM/NPM and Access switching (which is called “Aggregation” in this research note). It will take some time, starting with improved integration between APM and access switching, but it'll take place eventually.
Here are the reasons that this convergence will happen:
Effective deployment of network-based packet capture varies based on the ability to properly aggregate network traffic in key locations....The network must be designed with a traffic aggregation layer in place, and must use matrix or specialized network devices in order to get more granular filtering of traffic being sent to the monitoring systems....The traffic aggregation network can be leveraged not only for APM use cases, but also for security monitoring products and network performance monitoring tools.
So in order to make APM effective, products like Net Optics Director (for 1/10G networks) or Director xStream and xBalancer (for 10G networks) must be deployed. Technological convergence is the tendency for different technological systems to evolve towards performing similar tasks. Convergence can refer to previously separate technologies that now share resources and interact with each other synergistically. (see also wiki).
Modern applications and data centers introduce visibility issues for APM/NPM:
- Monitoring of encrypted traffic
- More complicated application delivery architecture
- Integration of on- premises and in-the-cloud applications (different locations which are also “elastic”)
Some vendors of network-based packet capture are counteracting those challenges by allowing packet capture and analysis on server-side devices (on-premises only). Those solutions has other limitations like performance challenges, need to support multiple OS and application servers as well as inability to monitor inter-vm traffic effectively.
The solution would be tighter integration of the visibility systems and APM/NPM products.
Your thoughts?
-- Sharon
Comments
Post new comment