>>> From: Don Gould <[email protected]> To: Craig Weinhold <[email protected]> Dat
> Putting flows in SQL is certainly doable (especially on a home network), but > be forewarned that network flows aren't as simple as you'd like them to be > and the sheer volume of indexing all that data is significant. That's why > most packages rely on an aggregate data store like rrd for fast min/max/avg > histories, and then keep a month or so of raw flow files for running detailed > reports against. > ta. yes, load is an issue... I'm working on a proof of concept. The 'home thing' really was just a use case to demonstrate an idea. Just my two cents but flow traffic itself even on a busy net only amounts to apx 1% of total bandwidth. Flows do not suck the life out of any equipment. Even under an attack some flows will get to your collector so you have a way to track most of what happened. bigfoot. _______________________________________________ Flow-tools mailing list [email protected] http://mailman.splintered.net/mailman/listinfo/flow-tools
