This is a case where the writer isn't in Brocade's target market or the article just hasn't aged well :) There are any number of open source router distributio

This is a case where the writer isn't in Brocade's target market or the article just hasn't aged well :) There are any number of open source router distributions using kernel-based forwarding (e.g. not fast enough for high PPS/limited utility in carrier environments). When it was introduced Vyatta was trying to sell hardware to compete with Cisco branch/ISR routers using the FOSS & make money on hardware+services model. Thing is Cisco ISRs integrate a lot more than routing (to the point of being a branch office in a box) so there wasn't a ton of demand for pricy commodity hardware. Imagestream was another vendor trying to complete on that model that has faded. Around that time NFV became the new hotness (moving services like NAT, VPN, firewall to VMs so big expensive routers can focus on just moving traffic). Now, kernel-based forwarding routers are glass cannons. They can achieve high data rates with large packet sizes but tend to fall over when exposed to high rates of small packets like you'd find in VoIP or DoS traffic. This limits their usefulness in environments where you need to handle and block a lot of bad traffic. Brocade acquires Vyatta and reworks it to use DPDK (userland forwarding) so it doesn't fall over. They aren't the only one: Juniper has vMX, Alcatel-Lucent has VSR. All router products designed to fit into a pretty specific niche--one where not selling hardware is the niche.