Fibre Channel (SAN)

Reply
Occasional Contributor
Posts: 12
Registered: ‎06-09-2010
Accepted Solution

48000 core risks

I have a large FC customer that has two 4/256 Directors running 6.4.1a that will be going EOS in less than two years. They are asking for any potential risks of keeping the directors in the core. Can any add or provide any experiences or thoughts around keeping these switches in the environment as core components?  

External Moderator
Posts: 4,974
Registered: ‎02-23-2004

Re: 48000 core risks

--->>>Can any add or provide any experiences or thoughts....

 

The advantage Brocade Director's vs other vendor i.e Cisco is follow:

 

Bracade have a rock solid backplane, and it is not longer a secret that Brocade have the best Backplane concent on the market.

 

I working since 15+ years with Brocade equipment, expecially with Directors since Brocade 12000, and I have a customer with a DUAL fabric builded with TWO 12000.

 

both work fine, we must replace here at one Time a FAN, and TWO Power Supply

 

In 10 YEARS !!!!!

 

I've myself in my LAB 3x 48000, One 24000, and One 12000, nothing was defective.

 

however you shlod keep in mind, in a critical environment, you should encourage the Customer to switch to newest technology like DCX.

 

 

 

 

TechHelp24
Valued Contributor
Posts: 761
Registered: ‎06-11-2010

Re: 48000 core risks

hi,

 

I agree with TechHelp24, I find 48K a very robust switch, which has provided my only a very small number of headaches compared to other models/vendors. But your customer has to bear in mind that problems can always occur and if the switch goes EOS, the time-to-resolve of an issue will probably increase due to the possible lack of FRUs or support maintenance contract, etc.

 

Rgds

Valued Contributor
Posts: 931
Registered: ‎12-30-2009

Re: 48000 core risks

I agree as well.

 

The focus for the customer should be the supportability of his environment as felipon suggested already.

Ie how much money is lost if the environment is not available for X amount of time.

 

EOS doesn't mean the gear will break one day later, but the older it gets the liklyhood of it breaking (in some form or shape) increases while the availablilty of parts and knowledge decreases.

 

 

Contributor
Posts: 22
Registered: ‎02-24-2014

Re: 48000 core risks

There are some risks. If you use over 8 individual ISLs instead of trunks to connect this to a switch with a newer ASIC generation, you could see SCSI resets on NPIV traffic that has to traverse the ISLs. It's a pretty rare configuration, but it happened to me.

 

If it's working fine now, just try to avoid doing new things with it once it's EOS and you should be ok.

Valued Contributor
Posts: 549
Registered: ‎03-20-2011

Re: 48000 core risks

agree with all of the above, BUT...
this also depends on what will be on the edges. if you'll stick a number of 16G switches in there, you will quickly find that good old 48ks wouldn't cope with this amount of pressure from the edges...

Join the Community

Get quick and easy access to valuable resource designed to help you manage your Brocade Network.