DirectAccess Server 2012 Behind a NAT with Teredo Tested

So here it is finally our REAL world test of Server 2012 DirectAccess.
We said REAL because and going off for a little rant here but stick with it as its quite important – We have read so many posts about people using DirectAccess 2012 and how great it is and how the IPHTTPs tunnel is on par with Teredo now, well no. Simply Teredo is still better and we have the proof to show you ( but in fareness we are the only people in the world who can test this behind a NAT). The difference is, is that we live in the real world and not test labs. We have over 5 year old servers we are installing this on. We have laptops that are not meant for even windows vista but we have worked endless hours getting Windows 7 working on them so that they just about work so we do not have to go begging for money that our bosses do not have just so we can use this technology. We do not have 10Gb/1Gb back bones from one test lab behind a NAT to another. We have 5-40mg broadband at the work place and if we are lucky 24mg at home (i’m not lucky only 13mg here 🙁 ). We live in the real world and this is why we have had a sharp increase in customers wanting the bolt on fix to help speed up their lives.
So little rant over lets get down to the tests.
Some notes: The ipv6-test.com is done on different equipment and broadband speed so don’t compare it to the last blog on speed we did under UAG.
Did you know Teredo is more secure? a link here on the benefits of Teredo here http://technet.microsoft.com/en-us/library/bb457011.aspx

First Test we did was through a 3g phone to see how long the initial connection took. We disabled Teredo, netsh int teredo set state disable and connected laptop to phone. This quite often seemed to take nearly 60secs to connect over IPHTTPs and browse the network. That means this could add 60secs to your boot time as it connects to the work place before anything else happens so that the machine can get its GPO’s. When we enabled Teredo netsh int teredo set state client it instantly connected.

The rest of the tests were at home via a 20mg fibre workplace connection and 13mg broadband connection on a latitude d600 11Mbs wireless card.

Second test was with ipv6-test.com. The first set of pictures shows browsing with Teredo were the second is IPHTTPS enabled.

That’s 1.25Mbs/s extra that we have to play with. We can already see IPHTTPS is slightly better than before but still way off par with Teredo.

Third Test was with downloading a 40768KB file from the work place straight to the laptop. The first set of pictures shows IPHTTPS downloading and the second is Teredo.

We noticed IPHTTPS was all over the place where as Teredo was Solid. Teredo was also 50KB/s faster. That sounds very little in performance. Again in the real world were your downloading real files like for instance a 1gb file (1/3 of a movie for example) you will see a dramatic time saving. Here is a small calculation on downloading 1gb over 50KB/s. 1 gb = 1024 mb and 1 mb = 1024 kb so 1024*1024=1048576 kb so you have to download 1048576 kb ok now 1048576/50=20971.52 second
20971.52/60=349.52 minute 349.52/60=5.82 hour
so you need about 6 hours to download it.

Our next blog is on DirectAccess Pandora’s box – What you don’t know about DirectAccess and how it could help your company save £££,£££’s

Conclusion – As we have said before we were very excited about Server 2012 when we heard about what it could do but the fact you can now get Teredo tunnels, well save everyone the hassle and time and buy the bolt on from just £250

Just email sales@mydanat.co.uk to say your interested.

Post a Comment

Your email is never published nor shared. Required fields are marked *

This blog is protected by Dave\'s Spam Karma 2: 23160 Spams eaten and counting...