Internet Protocol Flow Information Export (IPFIX) is an IETF working group. It was created from the need for a common, universal standard of export for Internet Protocol flow information from routers, probes, and other devices that is used by mediation systems, accounting/billing systems, and network management systems to facilitate services such as measurement, accounting, and billing. The IPFIX standard will define how IP flow information is to be formatted and transferred from an exporter to a collector. Previously many data network operators were relying on the proprietary Cisco Systems NetFlow standard for traffic flow information export.
The IPFIX standards requirements were outlined in the original RFC 3917. The working group chose Cisco NetFlow Version 9 as the basis for IPFIX. The working group submitted the IPFIX Protocol Specification to the IESG for approval in 2006.
Contents |
The following figure shows a typical architecture of information flow in an IPFIX architecture:
Metering,
Exporter IPFIX Collector
O--------------------------->O
|
| Observation Point
v
---- IP Traffic --->
A Metering Process collects data packets at an Observation Point, optionally filters them and aggregates information about these packets. Using the IPFIX protocol, an Exporter then sends this information to a Collector. Exporters and Collectors are in a many-to-many relationship: One Exporter can send data to many Collectors and one Collector can receive data from many Exporters.
Similar to the NetFlow Protocol, IPFIX considers a flow to be any number of packets observed in a specific timeslot and sharing a number of properties, e.g. "same source, same destination, same protocol". Using IPFIX, devices like routers can inform a central monitoring station about their view of a potentially larger network.
IPFIX is a push protocol, i.e. each sender will periodically send IPFIX messages to configured receivers without any interaction by the receiver.
The actual makeup of data in IPFIX messages is to a great extent up to the sender. IPFIX introduces the makeup of these messages to the receiver with the help of special Templates. The sender is also free to use user-defined data types in its messages, so the protocol is freely extensible and can adapt to different scenarios.
IPFIX prefers the Stream Control Transmission Protocol as its transport layer protocol, but also allows the use of the Transmission Control Protocol or User Datagram Protocol.
A simple information set sent via IPFIX might look like this:
Source Destination Packets ------------------------------------------ 192.168.0.201 192.168.0.1 235 192.168.0.202 192.168.0.1 42
This information set would be sent in the following IPFIX message:
| Bits 0..15 | Bits 16..31 |
|---|---|
| Version = 0x000a | Message Length = 64 Bytes |
| Export Timestamp = 2005-12-31 23:59:60 | |
| Sequence Number = 0 | |
| Observation Domain ID = 12345678 | |
| Set ID = 2 (Template) | Set Length = 20 Bytes |
| Template ID = 256 | Number of Fields = 3 |
| Typ = sourceIPv4Address | Field Length = 4 Bytes |
| Typ = destinationIPv4Address | Field Length = 4 Bytes |
| Typ = packetDeltaCount | Field Length = 4 Bytes |
| Set ID = 256 (Data Set using Template 256) |
Set Length = 28 Bytes |
| Record 1, Field 1 = 192.168.0.201 | |
| Record 1, Field 2 = 192.168.0.1 | |
| Record 1, Field 3 = 235 Packets | |
| Record 2, Field 1 = 192.168.0.202 | |
| Record 2, Field 2 = 192.168.0.1 | |
| Record 2, Field 3 = 42 Packets | |
As can be seen, the message contains the IPFIX header and two IPFIX Sets: One Template Set that introduces the build-up of the Data Set used, as well as one Data Set, which contains the actual data. Because the Template Set is buffered in Collectors it will not need to be transmitted in subsequent messages.