APPLICATION IDENTIFICATION AND QOS MONITORING
Troubleshooting complex networks requires the detailed knowledge of what applications are generating the traffic: some are more bandwidth or performance demanding, or are business critical, whereas some others are more tolerant to temporary performance impairements.
Efficient troubleshooting requires having all needed data at hand (bandwidth, delays, packet loss, sources, destinations and applications affected), and going “back in time” to observe differences before, during and after a network performance issue.
Network monitoring platforms must be designed to perform application classification, performance measurements, data storage in a flexible and scalable way. Otherwise, if they focus on one aspect only, organizations are forced to deploy multiple solutions, increasing the cost and complexity of network monitoring.
The need of application classification and of granular, continuous measurements risks to boost monitoring cost
troubleshooting complex networks
SmarTIE combines application traffic classification and performance monitoring functionality. Thanks to sophisticated decision intelligence, result of years of research and tuning, SmarTIE is capable of classifying the traffic generating applications via a lightweight inspection of the first few packets of a communication flow (e.g. after the first two packets only).
With respect to “traditional” DPI based approaches, this requires much less computation resources, and is thus suitable to be deployed in systems providing also other functionality.
SmarTIE combines application traffic classification and performance monitoring functionality adapting to existing computation resources
SmarTIE probes can thus be deployed as standalone servers, or agents in terminals, or as SW modules in existing servers or Virtual Machines, and be flexibly and dynamically distributed according to the monitoring needs of the network.
Each SmarTIE Probe reports its results to a hierarchical network of collection points, which make the results immediately available for operational purposes to the SmarTIE Management Station. Also plain NetFlow and IPFIX data coming from other routers or probes can be collected and analysed jointly with the more information rich data produced by SmarTIE probes.
Finally, SmarTIE can be enhanced with active tests, including the sophisticated traffic generation and emulation capability of DITGBox.
policy enforcement and QoS support
SmarTIE has a wide range of possible uses, including traffic and QoS surveillance, accounting, security and policy enforcement. In particular, its ability to achieve a reliable application classification after the first few packets of a communication flow (e.g. after the first two packets only) makes it possible to use it together with real-time policy enforcement points that either eliminate or throttle unwanted traffic.
Likewise, SmarTIE can recognise business critical and QoS demanding applications and pilot policy enforcement points to prioritize them on the fly
application aware planning
In Service Providers environments, SmarTIE is also suitable to be deployed in a distributed way even in edge devices like home routers, and is thus an ideal suite to conduct recurring campaigns to survey the performance of different access technologies, what applications (and how) are used by the customers, the QoS they receive. All this, of course, collecting and consolidating infomation in such a way that the privacy of end users is always preserved.
Likewise, in big Corporate environments or in Cloud/Hosting Providers, SmarTIE is suitable to be deployed in the more critical network datacenters, where it is paramount to control the QoS delivered to business critical applications, understand the remote VPN endpoints making more use of them, and timely plan for network bandwidth upgrades.
SmarTIE can detect unwanted applications and pilot policy enforcement points to remove or limit them in the network