[c-nsp] GigE cards for 6509

Tim Stevenson tstevens at cisco.com
Thu Oct 14 23:03:48 EDT 2004


I didn't explicitly state it, so want to add: the diff between the 
6148-GETX & the 6548-GETX is that the latter also has a dedicated 8G fabric 
connection in the presence of a fabric.

Otherwise, they are the same. Both are 8:1 oversubscribed, 6G max throughput.

Tim

At 07:57 PM 10/14/2004, Tim Stevenson stated:
>Yikes! Lots of misunderstandings here folks!
>
>Clarifying the architecture of 6148-GETX first of all. It is a classic 
>card, no fabric - so the 8G is not correct, you are perhaps thinking of 
>the fabric enabled (6548-GETX) version.
>
>However, both 6148-GETX & 6548-GETX are 8:1 oversubscribed, which means 
>you will NEVER get more than 6G out of these cards. There are 2 backplane 
>interface ASICs, which are each 4x1G chips, but only 3x1G is used on each. 
>Each 1G port on these chips connects to another ASIC (a mux so to speak) 
>servicing 8 front panel ports. So 8 front panel ports share 1G of 
>bandwidth, and there are 6 groups of 8 ports (contiguous, so 1-8, 9-16, etc).
>
>These cards are exclusively targeted for gig to the desktop/wiring closet 
>deployments, they are NOT intended for server connectivity or 
>high-performance data center deployments. They do not support jumbos, 
>don't support broadcast suppression, don't support distributed forwarding, 
>but do support optional IEEE & cisco inline power.
>
>As for the 6516, it has both a bus connection & an 8G dedicated fabric 
>connection (in the presence of a fabric). It has full access to the bus 
>(not internally oversubscribed), assuming no fabric and no other cards are 
>arbitrating for it. Across the fabric, it is 2:1 oversubscribed. The 
>"equivalent" "upgrade" to this card would be a 6724 or 6748 card, which 
>requires sup720.
>
>Tim
>
>
>At 03:15 PM 10/14/2004, cisco-nsp-request at puck.nether.net stated:
>>Message: 5
>>Date: Thu, 14 Oct 2004 15:37:26 -0500
>>From: "Church, Chuck" <cchurch at netcogov.com>
>>Subject: RE: [c-nsp] GigE cards for 6509
>>To: "Voll, Scott" <Scott.Voll at wesd.org>, "Paul Stewart"
>>      <pauls at nexicom.net>,     "Dave Breiland" <dave at opsource.net>
>>Cc: cisco-nsp at puck.nether.net
>>Message-ID: <16A2F11941294D469A25A7295F7A757303D49E4E at mspex01>
>>Content-Type: text/plain;     charset="us-ascii"
>>
>>I think the classic bus cards (61xx) are all 8GB full duplex.  Check CCO
>>for all the caveats involving SFMs, Sups, line cards, buses, etc.
>>Here's a decent link:
>>http://www.cisco.com/en/US/products/hw/switches/ps708/products_qanda_ite
>>m09186a0080159963.shtml
>>
>>
>>Chuck Church
>>Lead Design Engineer
>>CCIE #8776, MCNE, MCSE
>>Netco Government Services - Design & Implementation
>>1210 N. Parker Rd.
>>Greenville, SC 29609
>>Home office: 864-335-9473
>>Cell: 703-819-3495
>>cchurch at netcogov.com
>>PGP key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x4371A48D
>>
>>
>>-----Original Message-----
>>From: cisco-nsp-bounces at puck.nether.net
>>[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Voll, Scott
>>Sent: Thursday, October 14, 2004 4:23 PM
>>To: Paul Stewart; Dave Breiland
>>Cc: cisco-nsp at puck.nether.net
>>Subject: RE: [c-nsp] GigE cards for 6509
>>
>>I believe the back plain is 32Gig
>>
>>-----Original Message-----
>>From: cisco-nsp-bounces at puck.nether.net
>>[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Paul Stewart
>>Sent: Thursday, October 14, 2004 1:08 PM
>>To: Dave Breiland
>>Cc: cisco-nsp at puck.nether.net
>>Subject: RE: [c-nsp] GigE cards for 6509
>>
>>That's good to know.. anyone know the actual card throughput on the 6148
>>card?  Tried to find it online but haven't found anything useful yet...
>>this card (either the 16 port GigE or 48 port 10/100/1000) won't be
>>taxed at all but it's nice to know you won't run out neither..:)
>>
>>Paul
>>
>>
>>On Thu, 2004-10-14 at 16:05, Dave Breiland wrote:
>> > I am guessing it relates, but my experience is with the older GIG/
>>GBIC
>> > cards.  You could buy 8 port and 16 port versions.  The problem was
>>that
>> > they both had the same amount of ASICs.  As such, your ports will have
>>less
>> > dedicated bandwidth, on the 16 port version.
>> >
>> > Just my cents of knowledge...
>> >
>> > Dave
>> >
>> > -----Original Message-----
>> > From: cisco-nsp-bounces at puck.nether.net
>> > [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Paul Stewart
>> > Sent: Thursday, October 14, 2004 12:55 PM
>> > To: cisco-nsp at puck.nether.net
>> > Subject: [c-nsp] GigE cards for 6509
>> >
>> > We need some more GigE cards for our 6509.  I contacted our supplier
>> > today and was told that a WS-X6316-GE-TX 16 port copper GigE card
>>(which
>> > we already have one of and need another) is MORE money than a
>> > WS-X6148-GE-TX card which has 48 10/100/1000 ports on it.
>> >
>> > I'd be crazy not to buy a 6148 card it looks like... is there any
>> > catches that I should know of?  This seems too easy..;)
>> >
>> > We use these cards to connect servers but also run some basic QOS
>> > functions and Dot1Q trunking on some interfaces...
>> >
>> > Paul
>> >
>> >
>> > _______________________________________________
>> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
>> > https://puck.nether.net/mailman/listinfo/cisco-nsp
>> > archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>>_______________________________________________
>>cisco-nsp mailing list  cisco-nsp at puck.nether.net
>>https://puck.nether.net/mailman/listinfo/cisco-nsp
>>archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>>_______________________________________________
>>cisco-nsp mailing list  cisco-nsp at puck.nether.net
>>https://puck.nether.net/mailman/listinfo/cisco-nsp
>>archive at http://puck.nether.net/pipermail/cisco-nsp/
>
>
>
>Tim Stevenson, tstevens at cisco.com
>Routing & Switching CCIE #5561
>Technical Marketing Engineer, Catalyst 6500
>Cisco Systems, http://www.cisco.com
>IP Phone: 408-526-6759
>********************************************************
>The contents of this message may be *Cisco Confidential*
>and are intended for the specified recipients only.



Tim Stevenson, tstevens at cisco.com
Routing & Switching CCIE #5561
Technical Marketing Engineer, Catalyst 6500
Cisco Systems, http://www.cisco.com
IP Phone: 408-526-6759
********************************************************
The contents of this message may be *Cisco Confidential*
and are intended for the specified recipients only.



More information about the cisco-nsp mailing list