Specifications

DATA CENTER BEST PRACTICES
SAN Design and Best Practices 69 of 84
TOOLS FOR GATHERING DATA
Brocade SAN Health
Brocade SAN Health is a free tool that allows SAN administrators to securely capture, analyze, and report
comprehensive information about Brocade fabrics with switches running Brocade FOS and M-EOS operating
systems and Cisco MDS fabrics running SANOS/NXOS. It can perform tasks such as:
•Taking inventory of devices, switches, rmware versions, and SAN fabrics
•Capturing and displaying historical performance data
•Comparing zoning and switch congurations against best practices
•Assessing performance statistics and error conditions
•Producing detailed reports (in Microsoft Excel) and diagrams (in Microsoft Visio)
Note: In mainframe FICON environments, collect the Input/Output Configuration Program (IOCP) in plain-text format
(build I/O configuration statements from HCD), and upload the data. Brocade SAN Health matches the IOCP against
the RNID data.
Download Brocade SAN Health and nd details and instructions on how to use it at:
www.brocade.com/services-support/drivers-downloads/san-health-diagnostics/index.page
Power Calculator
Power savings is essentially a nancial issue, in terms of not only operational costs but additional costs of
upgrading power infrastructure due to growth. The power calculator can be downloaded from:
www.brocade.com/data-center-best-practices/competitive-information/power.page
Storage Traffic Patterns
Most storage arrays have tools for gathering port and LUN level performance data (contact the array vendor for
the appropriate tool). It is recommended that gathering a week’s worth of data will help in determining if the
there are enough resources to accommodate the new application requirements.
The data should reect both normal and high utilization, such as data that reects the end of a quarter.
The metrics to collect are as follows:
•Percent of reads
•MB/s reads
•Percent of writes
•MB/s writes
•Worst-case latency (ms)
•Number of SCSI commands/second
•Cache hits
•Queue depth
Server Traffic Patterns
On the server side, there are Windows and Unix tools for collecting CPU, memory, and network utilization built
into the OS. HBA vendors also provide tools to gather the following on a per-port basis:
•Percent of reads
•MB/s reads
•Percent of writes