Our PFI partner is under the impression that they will need to rip all the cat 5e cabling out of their 8 year old building and replace it with cat6 in time for the BSF hand over in 2012. The person stating this couldn't tell me what the difference would be but I don't think it was him that made the decision either.
As this will no doubt cost the school is there an arguement to support this?
We have fibre between the cabinets and most desktop machines are still only 100Mb/s. I accept that data use will go up but I don't anticipate it going up that much that 1Gb/s to the desktop will be an issue and IIRC cat6 only supports speeds greater than this over short runs.
Even at the end of the first BSF contract (approx 8 years from now (IIRC its a 5 year contract and we get handed over in 3 years)) I don't think 1Gb/s to the desktop will be limiting based on our progress over the last 8 years and everyone's predictions that everything is going wireless to personal learning devices.
Please let me know your comments.
Polite ones, so that I can use them to support my argument that this replacement is not necessary.
To be honest you've answered your own question really, you will notice no difference. CAT5e is capable of operating at 1000Mbps.
I can only come to the following conclusions -
1. Not all your infrastructure is CAT5e but only CAT5.
2. There are areas of your network where the installation is poor and re-wiring may improve performance. If you twist or bend cables too much this can reduce throughput.
In my experience installing better (typically managed) branded switch equipment can improve network speeds drastically. Using the same brand of network equipment throughout your network can also reduce problems.
If all the above is already checked and OK in your opinion; then additional capacity at the core of the network (supporting 10Gbps) may be the way forward, if your existing infrastructure is capable of 1000Mbps.
As stated above Cat5e is perfectly able to support 1GBs, we install Cat6 on new installs because we have to (MoE directive) but would not rip out perfectly good 5e, 5 perhapse but 5e is fine. Cat6 has the benifit of being more robust thanks to being thicker and it also has stricter guidelines for installation than 5e that could just be dragged about the site with reacless abandon.
Category 5e copper twisted pair (Cat5e) / BS EN 50173 Class D
Throughout this document we have referred to Cat5e and Cat6 cabling because these are the more recognised terms. The equivalent British Standards are BS EN 50173 Class D and Class E respectively. Pros: Less expensive than Cat6. Can cope with a 1Gbps data rate. Able to carry Power over Ethernet with no complications. Compatible with 802.3ab.
Cons: Maximum 100m run of cable without a repeater (maximum of two repeaters).
Category 6 copper twisted pair (Cat6) / BS EN 50173 Class E
Pros: It can handle higher frequencies than Cat5e and has superior transmission qualities but, in practice, will not support any Ethernet standard or data rate not supported by Cat5e.
Cons: More expensive than Cat5e. Maximum 90m run of cable plus 10m for patch cords. Because of thickness of cable, there are limits to how tightly it can be routed in bends that might cause routing issues.
There is no requirements whatsoever for cat6 in new school builds
The network shall be cabled with fibre optic cable or Cat5e or Cat6 copper cabling.
Installed cable shall have the ability to support data rates of up to 1Gbps.
Where fibre is used for longer spans, over 550 m, it should be single-mode.
Where single mode fibre is used it should be OS1, 8.3/125 micron fibre.
Where multimode fibre is used it should be OM3, 50/125 micron fibre.
Cat3 shall be replaced with Cat5e or Cat6, and fibre optic cable where required.
Cat5 should be replaced with Cat5e or Cat6, and fibre optic cable where required.
I think the other important factor to include is CAT6 is considerably more expensive than CAT5e. I would put this down primarily to manufacturing costs, but also installation requirements require more careful consideration (due to increased thickness).
It all comes down to cost really more than anything. To you give you an idea, 300 metres of CAT6a cable is around £400, whereas 300 metres of CAT5e is around £60.
yes, but the profit margins are higher from a contractors point of view - both in terms of supplying the cable and the time required to lay it. PfI partners have a vested interest in supplying the 'best quality' to meet 'future needs'
If you're connecting cabinets, surely you'd be using fibre? The cost will be similar (assuming same contractor is used) and 10 years from now when generations of idiotic electricians have run electrical cable through the conduits clearly marked "data only", performance will be unaffected.
yes, but the profit margins are higher from a contractors point of view
That's not necessarily always the case. Larger profit margins would be easily achieved rolling out CAT5e. Your overheads of CAT6a are far greater (over 3 times) using the numbers I quoted above and that's before quoting labour and other parts required (such as trunking).
I'm all for future proofing, but I really cannot see the need or justification rolling out CAT6a. Something like 90% of networks are using CAT5 or CAT5e and both of these are capable of 1000Mbps. CAT5 just isn't officially recognised whereas CAT5e is.
From another perspective, have you searched for 10GbE switches? It just goes to show just how far away from mainstream the technology is due to prices Intel 10GbE adapters are around the $999 at the moment.
The technology of 1000mbit has matured now and all our recent switch purchases will do 1gb to the desktop over our Cat5 and cat5e. In a few cases I've had to clip the ends and terminate it again but thats down to a bad install or wear.
I have every expectation that in a decade somone will have figured out how to put 10gb down Cat5e and if nothing else you can tie 2 or 4 Cat5e together for a trunk line very cheaply.