In some cases, the main reason in my opinion is only commercial...
I'm not sure what to think.
Astaro's a great product and I stand by it (I convinced our local IT vendor to carry Astaro) but perhaps someone needs to look at the specs sheet. Lab results are great but when 'real world' performance (according to several Astaro Partners with kit in the field and skin in the game) sits at around half what the sales literature claims, something needs to be reviewed.
Case in point, IPS throughput is rated(on v7) around 100Mbps(v3 h/w) and VPN at around the same. We tuned IPS for our network and it was running fine till I got a complaint the internet was down. Traced the problem down to a user who was trying to dump 4GB across a 6Mbps VPN. At 10% of rated capacity on both sub-systems, the little brick shouldn't have any issues.
Obviously you can't turn *everything* on and expect the unit to meet it's advertised specs but when you're on the bottom end of the unit's recommended user count, running off a 'Net connection limited to 20% of what the UTM is rated for, you shouldn't run up against the hardware limitations.
All this said, barring any drastic changes over the next two years, I'm still sticking with Astaro when our licenses are up for renewal. Will probably move towards software however as we have shiny new ESXi servers and I've worked out some of the bugs with Astaro in a VM environment.[:)]
According to RAM extension - that's why I'm big fan of software/virtual version. From the other hand, I understand Astaro also. See that all those users with 120, when extending memory can handle more users - instead of migrating to 220.
According to RAM extension - that's why I'm big fan of software/virtual version. From the other hand, I understand Astaro also. See that all those users with 120, when extending memory can handle more users - instead of migrating to 220.
You are right, but the CPU of 120 is not the same of a 220...
