Switching in the Home
I need to bump up the switching at home.
Right now it's not all that fancy, as I'm just moving into the gigabit age at home (I know, I know). I've got the expected gaggle of unmanaged crap switches, and they're performing well. The planet GSD-803 performs the best of all, I think, as it doesn't need the monthly reset like the D-Link and Netgear unmanaged switches do. But enough of that. I want to get Gbit all the way into the home, and I need at least one managed switch.
I was looking at two cheapo switches, the HP1810G-8 and the Netgear GS108T-100NAS . Considering the ridiculous SC problems I've been having with the HP1810G-24 I'm using for the Atlantis ESX cluster,
Why would I consider buying it? Because the Netgear apparently has the same damned problem. Is it a perception issue, that we are somehow wrong to expect the dinky-toy management service console of this product to actually perform as designed? Do they get a free ride on quality control in that aspect, simply because they're merely a $100 switch?
Why not just grab a Netgear 3500L router, shut off the wireless (or use it; your choice) and use its massively expended capabilities, thanks to OpenWRT (white russian fo' life) and the ability of linux to set up multiple vLANs at will. It seems that, if I can stand the loss of the ports or foot the cost of another crap gigabit router (where I get failing hardware for 1/3 the price) it's less of a loss.
Right now it's not all that fancy, as I'm just moving into the gigabit age at home (I know, I know). I've got the expected gaggle of unmanaged crap switches, and they're performing well. The planet GSD-803 performs the best of all, I think, as it doesn't need the monthly reset like the D-Link and Netgear unmanaged switches do. But enough of that. I want to get Gbit all the way into the home, and I need at least one managed switch.
I was looking at two cheapo switches, the HP1810G-8 and the Netgear GS108T-100NAS . Considering the ridiculous SC problems I've been having with the HP1810G-24 I'm using for the Atlantis ESX cluster,
I mean, really, even in the lower-end enterprise space, what kind of load-testing wasn't done on this thing that let a problem like "oh, the SC freakin' bails all the time under load" get through? Why is it not recalled or replaced by a better-tooled device? But I know why:... I'm still considering buying their even-lower-end 8-port model. It does fewer vLANs, it does slower switching and it probably crashes on the SC even more often than the currently-floptastic HP1810G-24 unit.
Why would I consider buying it? Because the Netgear apparently has the same damned problem. Is it a perception issue, that we are somehow wrong to expect the dinky-toy management service console of this product to actually perform as designed? Do they get a free ride on quality control in that aspect, simply because they're merely a $100 switch?
Why not just grab a Netgear 3500L router, shut off the wireless (or use it; your choice) and use its massively expended capabilities, thanks to OpenWRT (white russian fo' life) and the ability of linux to set up multiple vLANs at will. It seems that, if I can stand the loss of the ports or foot the cost of another crap gigabit router (where I get failing hardware for 1/3 the price) it's less of a loss.
Labels: HP1810G-24, HP1810G-8, Netgear3500L, openwrt, WhiteRussian, wrt54gl
<< Home