As we all know, network infrastructure is built upon a combination of Switches, routers, Gateways etc. This is what enables us to transmit data from one point to another.
With the growth of SDN and products such as Cisco ACI and VMware NSX, Application-Centric Networking has started to come up in the forefront. Application Centric Networking essentially, is the networking based on Application. Here, the network administrator would manage a specific application set rather than servers, routers and switches.
It is still important to measure latencies, packet drops and other core networking elements, however with Application Centric Networking, these are not the main success criteria.
So, what determines success in Application Centric Networking?
Success is measured in the form of End-User Experience and Application Response times. It doesn’t matter how low your latency, but if your end-user keeps complaining of application slowdowns, its still an issue.
End User experience is a key success determinant in Application Centric Networking. To measure End-User experience, the key metrics are Application Response Times and service/server uptimes.
There are several questions you may need to ask –
- Can I see how Applications communicate with one another?
- For multi-tiered applications do you know where the bottlenecks could be? (In the web-tier or the Database Tier?)
- What kind of Data and how much Data is Flowing between them?
- Moreover, am I getting affected because of some underlying Infrastructure Issue? (CPU?, Memory?, Storage?)
- Are there any congestions with the Top-Of-Rack Switches?
The top challenges that Networking teams face are alert fatigue (emails or red blinking lights on the screen) from their network monitoring tool. With them being so busy, how do they focus on issues that truly matter to them and the organization. With an application-centric approach, teams can focus on the networking challenges that are directly impacting the application performance or are behind the disruption. So for example, you may care about your billing or checkout application slowdown due to a congested network port more than other applications.
To answer some of these questions, you would need the help of Application Dependency Maps, that can answer all the questions above by telling you where exactly the bottleneck and how it impacts your End User Experience.
A tool like Uila can create automated Application Dependencies for your environment while correlating with infrastructure and Network devices to give you a complete picture. Directly from the Application Dependency Map, you know if the associated port is congested or facing discards or errors due to a cabling issue.
- Mitigating Risks at a Datacenter during a crisis
- VDI Tips and Tricks for Desktop Professionals #3: Network Monitoring
- VDI Tips and Tricks for Desktop Professionals #2: End-to-end VDI and Application Visibility
- How the IT world has changed in the last week?
- VDI Tips and Tricks for Desktop Professionals #1: Resource Provisioning
- What's new in Uila 3.3
- Importance of Application-Centric Network monitoring
- Importance of Application Anomaly Detection
- Storage issues caused due to VM stunning
- Load Balancers - What can go Wrong?