2 ms·
No. Netflow makes it very difficult to capture tail latencies, and outliers. Netflow's aggregate model makes it very hard to detect things like buffers falling
by sargun 9y ago
No. Netflow makes it very difficult to capture tail latencies, and outliers. Netflow's aggregate model makes it very hard to detect things like buffers falling over, and making intelligent decisions based on that.
- youdontknowtho 9y agoThere are also on-box perf counters that can be read via SNMP... My point is that a networking department for a wall street firm would know that there were more tools available than ping and traceroute.