Archive for Testing

Validating Application Detection Signatures

In the new world of next-generation networks, pretty much every leading network equipment manufacturer (NEM) today has application-awareness built into their products. Whether it’s an application firewall, serving gateway or edge router, they’re all using deep packet inspection (DPI) to look deep into the network traffic to identify the specific application.

For example, Cisco has Application Visibility & Control, Juniper has AppSecure, Palo Alto Networks has App-ID, Sandvine has Traffic Identification and Tellabs has Application Identification.

Each vendor has their own proprietary database comprised of hundreds or thousands of application signatures and on finding a match, their system can then take action based on the defined policy (e.g. block an application, apply QoS, etc…)

Before these new application signatures are released however, testing is needed to ensure the accuracy of the detection. One of the major challenges is to avoid the false positive, in which an application is misclassified.

Full Post »

Bookmark and Share

Mu App Quadrant #3 – Skype Voice on Mac (OS X) Expends at Least 28% More for Consumers and Operator Networks than on Other Devices

In the previous versions of the Mu App Quadrant we first compared the most popular video services, and then specifically focused on Netflix across multiple endpoints. Now, in this third edition we have looked at Skype. Not only is it one of the most popular voice apps on the internet with over 30 million users online at peak times, it’s also one of the biggest bandwidth guzzlers, due in part to the P2P nature of its architecture.

Full Post »

Bookmark and Share

Driving Real Application Traffic Through Junosphere Virtual Infrastructure

Today, Juniper announced Junosphere™ Lab, an innovative on-demand service that gives service providers and enterprises immediate and low cost access to a virtualized environment for designing and testing networks. Very cool stuff – leveraging the power of the cloud and helping customers dramatically reduce their TCO while accelerating the time to model networks.

Real Traffic in a Virtual Environment

So when you spin up a network environment and model a production topology, you’ll then need a way to create realistic application traffic to understand its impact across the network. That’s where we come in.

Mu Studio Performance has been integrated into the Junosphere Lab so you can just as easily spin up (and tear down) virtual instances of our performance testing solution to quickly and accurately recreate a mix of applications that represent the production environment – that is, real users on real devices, running real applications.

Full Post »

Bookmark and Share

Ensuring the Accuracy of the Mu TestCloud Application Tests

In a previous blog I discussed how we had started to build out the test content for different kinds of applications across categories like P2P, video, chat and social media in our Mu TestCloud store. Fast-forward to today, and we’ve now got well over 2,000 tests, with coverage for hundreds of different apps. We’ve also got lots of customers who are actively using these ready-to-run tests for a wide range of use-cases – everything from verifying application detection signatures to validating application policies, as well as billing and charging.

But regardless of their domain, there are two common questions that customers are curious to understand:

1. How do we select the applications in the first place?
2. How do we ensure the accuracy of the tests?

So for this blog I’m going to give you a behind-the-scenes view into our test content creation process.

Full Post »

Bookmark and Share

Mu App Quadrant #2 – Netflix on iPhone & iPad Impacts Operator Networks More than on Android Devices

In our first edition of the Mu App Quadrant, we compared the run-time aspects of the most popular desktop video applications – Netflix, YouTube, Hulu and Amazon Video on Demand – and showed that Netflix is not particularly friendly to both Consumers and Operators.

It’s not just the apps – devices matter too!
Operators continue to struggle with the unpredictable growth of applications and the devices used to access them. With millions of people running applications like Netflix, spikes can occur on the network which often leads to disruption of other applications and services.

Recently, Korea Telecom suffered a network outage where a third-party app took the voice-call success rate down to a mere 10% because the signaling traffic generated by the app overloaded its network.
Full Post »

Bookmark and Share

Creating an Application Mix to Model the Production Network

Video and P2P Rule!
The traffic making up today’s networks is in a rapid state of flux. Just last week Sandvine, in their Spring 2011 Global Phenomena Report, noted that real-time entertainment continues to increase, and within North America represents almost 50% of peak fixed access traffic (much of this of course is due to Netflix). P2P traffic also continues to carve out a sizeable piece of the pie at around 20%. The rest is a mix of voice, business apps, games, Facebook and chat.

What’s interesting though is that the relative amount of traffic that isn’t application-level is tiny – all the stuff that makes networks run like DNS, ICMP, BGP and so on.

Full Post »

Bookmark and Share

blitz.io: How many dynos do I need on Heroku?

Whatever PaaS offering you are using, this is a very common capacity-planning question: How many dynos/instances of my app do I really need in order to support x concurrent users at any given time. Auto-scaling and elastic-load-balancers are awesome, but you still need to know what you are up against. With the ruby gem from blitz.io, this is super easy to iterate and find out for yourself before you go live!

Full Post »

Bookmark and Share

Splits, handshakes and bananas

It doesn’t matter what business you are in, but #$*(‘ing with your customers by releasing advisories and threatening them to buy your product or services is just plain dumb. For those that are following the TCP-split-handshake epic #fail saga, I have to say, the vulnerability itself is a clever hack. By using double-SYN’s or simultaneous connections (which is incredibly rare and non-existent on most modern networks), one can trick firewalls and IPS’ in not tracking state. This is reminiscent of the original classic Insertion, Evasion and Denial of Service that covered lots of grounds in the pitfalls of deconstructing application state in real-time completely based on the packets that are flowing through.

Full Post »

Bookmark and Share

Mu App Quadrant #1 – Not all video streaming apps are created equal…

There’s been a lot of debate recently about the impact of video apps on the network. According to Nielsen, Netflix alone now accounts for 20% of downstream traffic during peak times in the United States. In a previous blog Kowsik explained the behind-the-scenes interactions that are happening unbeknownst to you when you watch a Netflix movie. So that got us thinking – are all the popular video apps as network-intensive as Netflix? Are some video apps more user-friendly with their bandwidth consumption than others? Are some video apps more operator-friendly with their consumption of networking resources?

Full Post »

Bookmark and Share

Mommy, Netflix is eating my firewall!

Personally, as a consumer, I love Netflix, but it hasn’t been the darling of service providers and ISP’s lately. You can read about the Canadian ISP saga here. Our imminent next release of Mu Studio will enable our customers to recreate 1,000,000 concurrent Netflix users watching a movie, so they can understand the impact of their application aware networks. One thing is pretty clear: compared to YouTube, Netflix inflicts so much more pain on the network. Credit for this blog goes to Yuri who did all the reverse engineering. And he’s signed up to Netflix to watch movies during work for “research” purposes. :)

Full Post »

Bookmark and Share