Posted 01 January 2011 - 11:34 AM
Posted 02 January 2011 - 09:31 AM
Good post and responses everyone. I think I learned a lot. Based on this discussion, I believe my switch (link below) is a good switch?
NETGEAR JGS516 10/100/1000Mbps Gigabit Rackmount Switch 16 x RJ45 8,000 MAC Address Table 2MB Buffer Memory
http://www.newegg.com/Product/Product.aspx?Item=N82E16833122057&Tpk=JGS516
Main Server - WHS 2011, Core I5-2500, 12T RAID 5 (5x3T) + 2T of Mirror + 2T of backup
Second Server - 2008R2, Core I5-2500, 12T RAID 5
Main Systems - Core I7-2600k, 16 Gigs DDR3-1600, 180 Gig Intel 330 SSD Max IOPS 240 Gig Vertex 3, 2T Sata 3 for local Backup
Other systems - Core I7-2600, Core I3-530's, Core I5-2500, Core I7-920, Core I3-2100, and G620 (see System List)
My Blogs - The Docs Blog and Tablet Resource
BYOB Videos - TheBYOBPodcast
For a complete system List: Computer Systems
Posted 02 January 2011 - 01:40 PM
This seems to be a good switch, however it has a fan which can make a bit noisy for home environments depending on your setup. The dlink greens are fan-less and absolutely quiet.
Posted 14 January 2011 - 04:25 AM
Best practice would be to have a wall-plate with as many jacks as are needed (plus some spares). What actually happens would probably involve a local switch.
Cable is cheap and you can get up to 6 jacks on a single wallplate, so I'd try and go that route if I were you, using the patch panel that you have. (Remember though that to get cat6 speeds/reliability, everything has to support cat6, from the cable to patch panel to the connectors/terminations.)
Posted 14 January 2011 - 06:51 AM
Posted 14 January 2011 - 08:52 AM
Agreed. If you have the time and money you could run 4 runs (at least you only have to do 1 pull and then pull all 4 together (if you are using conduit)), but not always practical. Just remember that all 4 devices will be limited by the interconnect, which in your case probably won't be a problem.That is ideal, but it may not be practical.
Posted 14 January 2011 - 08:53 AM
Posted 14 January 2011 - 12:34 PM
Posted 05 February 2011 - 11:07 AM
So what's up with ports dying? Why is that? Low quality hardware?You might consider getting more ports, so as they die and they will, you do not have to get an entirely new switch. I have a 24port and use about 16 of them. It also leaves room for expansion and the additional cost is minimal.
Posted 06 February 2011 - 12:07 AM
Main Server - WHS 2011, Core I5-2500, 12T RAID 5 (5x3T) + 2T of Mirror + 2T of backup
Second Server - 2008R2, Core I5-2500, 12T RAID 5
Main Systems - Core I7-2600k, 16 Gigs DDR3-1600, 180 Gig Intel 330 SSD Max IOPS 240 Gig Vertex 3, 2T Sata 3 for local Backup
Other systems - Core I7-2600, Core I3-530's, Core I5-2500, Core I7-920, Core I3-2100, and G620 (see System List)
My Blogs - The Docs Blog and Tablet Resource
BYOB Videos - TheBYOBPodcast
For a complete system List: Computer Systems
Posted 06 February 2011 - 02:52 AM
Expansion room is good, but dying????? If I loose a port (which has not yet happend) I will want to replace the whole unit. He does make a good point on expansion though.
Posted 06 February 2011 - 03:30 AM
I have never had a port die on my home network. We have had bad ports at work but this is usually due to a lightning strike or surge and the switch acted like the fuse.Ports dying sounds a bit odd to me too.. Concidering what i just read from the specsheet of a d-link 16-port gbit router.. If it dies according to MTBF figure, you plug it in today and it dies in july 2105. Which makes me wonder how in the heck have they come to that figure anyway..
Posted 19 March 2011 - 07:58 AM
I can confirm that ports do sometimes die; not often, but it does happen, even on expensive hardware. From what I've heard from the manufacturers it's mostly the driver circuit for the chip, but I'm no EE so I can't say definitely. We've had it happen to a few ports at work - we usually replace the switch cause it's too much of a hassle to keep track of the dead ones, and also cause it doesn't happen often so it doesn't cost too much.Ports dying sounds a bit odd to me too.. Concidering what i just read from the specsheet of a d-link 16-port gbit router.. If it dies according to MTBF figure, you plug it in today and it dies in july 2105. Which makes me wonder how in the heck have they come to that figure anyway..
If at first you don't succeed, do it like your mother told you.
0 members, 0 guests, 0 anonymous users