The up.time IT Systems Management Blog

Posts Tagged ‘NetFlow’

NetFlow: Network analytics at your fingertips

Wednesday, August 11th, 2010

In my last post I mentioned I would talk about a few topics on a more technical level so today I’ll be talking about our NetFlow integration with a product named Scrutinizer from Plixer. I’ve also created a video introducing the NetFlow integration here.

Would you like the ability to zoom out to the 40,000 foot level and look at how your SLA’s are performing, yet drill down to have a look at who are your worst offending users and applications on your network? How about we take it one step further and have a look at who’s using up the most amount of bandwidth on an individual port on your core switch? Would you believe me if I said you can get all that power and visibility from one tool in the palm of your hand (if you have a mobile with a decent web browser)? Alright, that’s enough from the sales part of me. Yes, you do get all of the above with up.time.

It’s easy to access the NetFlow section from within up.time. By clicking on a server that is on the same network as a NetFlow monitored switch you will see a link in the Graphing section for NetFlow metrics. This gives you direct visibility into the network usage of the server and the top networking applications; and that’s just the default NetFlow view. From there you can slice-and-dice all of the network metrics and flows and find out the top applications, highest bandwidth usage, communication flows, and many other detailed network metrics coming from the server. It provides you with almost infinite drill-down capabilities into detailed low-level network metrics for your network administrators.

The GlobalScan dashboard gives you visibility into your global infrastructure availability, and now we also provide detailed visibility into your network usage and availability as well. You get detailed network analytics without having to drill down into a complicated profiler tool, all from within the up.time interface.

For example: it will show you which server is using up the most amount of bandwidth; and if it was a rogue system serving torrents and saturating your network pipeline up.time would bring that to the surface and show you immediately without requiring hundreds of clicks to get at that data.

With NetFlow Analytics you get higher visibility into what is happening on your network without having to sit there watching the packets fly by and this helps your network team proactively trace patterns before outages can occur. For a look at how this all looks and works feel free to check out our video on NetFlow – click here to check it out.

Tags: , ,
Posted in NetFlow Monitoring, Systems Management | No Comments »

Netflow | Network Data You Can Actually Use

Monday, June 21st, 2010

In the latest release of up.time there are a whole bunch of goodies, but I’m going to take a moment to talk about one specific feature that you should all be aware about, our Netflow capabilities.

To understand Netflow, let’s first start with how we (as in anyone who cares about the network) would do diagnostics and troubleshooting any kind of network error in the past. Trust me, I used to do this as a junior network engineer to pay my way through school, and I know all too well the pain of what I describe below.

First, you would go to your console tools, run those, and examine point stats. Because the point stats are “just the state of the network at the point in time”, and because the consoles are so hard to navigate by text, trying to find out trouble areas or diagnose distributed outages would be like finding a “needle in a haystack”.

Second, ok we’ve figured out that the point in time stats are a pain, if we’ve figured anything out, maybe we would now switch to something with some historical stats like MRTG or Cricket. The bad thing about these tools, they display static graphs over fixed time periods, trying to do any ad-hoc analysis is impossible. Even worse, these tools sort of illuminate key metrics like throughput, capacity utilization, dropped frames, but once you get to that level of figuring out which connections are being flogged, what do you do next? You have zero visibility into the traffic, the applications or any context over why that port or set of ports is underperforming or saturated.

Third, we bring out the BFG (Big !@#$%^& GUN). The passive network profiling tool. You configure the spanning port on the switch or router, you hook up a big honking expensive passive network profiler and what do you do? You WAIT. You wait for this network data to get collected, then you spend hours pouring over the ultra granular network profiling data in hopes of figuring out what is happing on your network.

Does the above sound like sanity or insanity? By the time you deploy the BFG, maybe your users are done watching the world cup viral videos, or maybe the DDOS attack is over.

Let me tell you what you need instead:

You need to be able to alerted on general network outages and performance issues in as they occur
You need to be able to drill down into network traffic breakdowns on network devices AND servers
You need to be able to pro-actively have a network dashboard that focuses less on profiling type operations, and more on network threats, misconfigurations and common applications that cause trouble on the network
You need to be able to perform ad-hoc analysis at will, on demand to rapidly gain insight into what’s going on.

This is what up.time’s netflow capabilities provide. If you want to see more, join one of our what’s new webinars and I’ll be happy to take you on a tour. Click here to register.

Here’s some eye candy to whet your appetite:

up.time NetFlow Monitoring up.time NetFlow Dashboard

Tags: , , ,
Posted in Systems Management, uptime | No Comments »