Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

4923

February 23rd, 2017 06:00

Cisco DCNM vs Brocade BNA and other tools

Hello users and Experts,

One of our big customer request that we evaluate the difference between Cisco and brocade offer about Monitoring tools and Advanced feature. Actually, they use Brocade Gen 5 Director and BNA with MAPS and they would to know if advanced function ant tools are better from Cisco (think about Slow Drain Device, monitoring bandwith, Buffer utilization, performance, diagnostic and troubleshooting, etc..)


Do you have some customer experience, personal experience and documentation about that please ?


Thanks for you help.

Regards,

Mat

38 Posts

February 23rd, 2017 08:00

I would like pointing you to a Cisco customer reference story - a migration to Connectrix MDS 9700's / MDS 9148S's: http://www.cisco.com/c/dam/en/us/products/collateral/storage-networking/mds-9700-series-multilayer-directors/manufacture…

We used the DCNM slow drain analysis to identify some bottlenecks without much effort. The tool shows were and when Buffer to Zero conditions are occurring and the related drop and reset counters, too. Critical values are highlighted "red" or "yellow" - I have never seen a better tool!

We increased buffer credits between UCS fabric interconnects and the MDS 9700's and deployed F-port channels. End device ports - providing up to 100 B2B's - got more credits on the switch side for performance, too.

After that, no high B2B to zero conditions were visible.

Some MDS highlights are: Port monitoring will be pre-configured. Frames having errors will be dropped - no defect frame will be forwarded to unsettle any end device. If required, ports will be error disabled. Input- / output queuing will prevent from running into head of line blocking situations. The crossbar architecture will guarantee equal latency and high throughput for all I/Os.

The customer feedback is:

  • ..."the entire changeover process to the Cisco MDS solutions went smoothly without disruption. The only difference users noticed was improved performance."
  • “We no longer experience errors or unbalanced links. Every link to the SAN fabric carries the same bandwidth, and all deliver outstanding performance.”
  • "With Cisco Prime Data Center Network Manager, Krones also has enhanced visibility into the SAN to identify and remedy bottlenecks, enhance link utilization, and analyze events weekly to optimize performance.

Last but not least: DCNM is able to discover vSphere, Dell EMC and other vendor storage through SMI-S providing end to end monitoring capabilities from a VM to a single LUN. 

I hope this customer experience will help.

Best regards,

Swen

15 Posts

February 23rd, 2017 23:00

Thank you Swen. Very interesting.

About Slow Drain device detection, the Cisco switches that you mentionned support the SDD detection.

We have help our customer with CMCNE to identify BB0 but SDD detection was not possible because they don't use the right code (available from FOS 7.4).I know that Brocade BNA (CMCNE) could help customer like DCNM about that.

Does DCNM could help customer to monitor the SAN load with IOPS or frame (bandwith load is available on DCNM and BNA) ?

38 Posts

February 24th, 2017 01:00

All MDS 9000 will provide more or less SDD features. SDD has been improved over time by moving features from software to hardware and by adding firmware extensions...

On YouTube, there are some nice videos about DCNM - SDD:

...or the short version:

There are more good MDS videos, like:

You will find all performance related features within the DCNM fundamentals guide:

I hope this will help.

Best regards,

Swen

No Events found!

Top