Brocade Communications Systems Network Advisor 12.3.0, Руководство пользователя

Страница: 1190 / 1940

1138 Brocade Network Advisor SAN User Manual
53-1003154-01
Flow Monitor
29
Flow Monitor
Flow Monitor enables you to monitor all the traffic passing through E_Ports, EX_Ports, F_Ports, and
XISL_Ports using hardware-supported flow parameters. It also lets you define your own monitoring
flows using combinations of ingress or egress ports, source and destination devices, logical unit
numbers (LUNs), and frame types to create a flow definition for a specific use case.
Flow Monitor provides support for monitoring the following flows and traffic:
•
Learned and static flows for traffic passing through E_Ports and F_Ports
•
Learned and static flows monitoring edge-to-edge traffic, edge-to-backbone traffic, and
backbone-to-edge traffic passing through EX_Ports
•
Learned and static flows monitoring traffic inside logical fabrics and inter-fabric (routed) traffic
passing through XISL_Ports
•
Learned and static flows monitoring inter-fabric traffic and backbone traffic passing through
backbone E_Ports
In Fabric OS 7.1.x and earlier, the Advanced Performance Monitor (APM) provided the following
monitors: End-to-End, Frame-based, ISL, and Top Talker. In Fabric OS 7.3.0, Flow Monitor provides
you with the following abilities in addition to those provided by the APM:
•
Monitoring of application flows (for example, a flow within a fabric from a host to a target or
LUN) at a specified port.
•
Comprehensive visibility into application flows in a fabric, including the ability to learn
(discover) flows automatically.
•
When N_Port ID Virtualization (NPIV) is used on the host, you can monitor Virtual Machine
(VM)-to-LUN level performance.
•
Capturing statistics for specified flows, which provides insights into application performance.
These statistics include transmitted and received frame counts, transmitted and received
frame throughput rates, SCSI Read and SCSI Write frame counts, the number of SCSI Reads
and Writes per second, as well as others.
A sample use case would be to monitor throughput statistics for inbound traffic between a
source device and a destination device. For an example procedure for this use case, refer to
“Monitoring LUN level statistics” on page 1153".
•
Monitoring of various frame types at a switch port to provide deeper insights into storage I/O
access patterns at a LUN, reservation conflicts, and I/O errors. Examples of the frame types
that can be monitored include SCSI Aborts, SCSI Read, SCSI Write, SCSI Reserve, all rejected
frames, and many others. For a list and description of the frame types that can be monitored,
refer to
“Flow frame type parameters” on page 1129".
•
SCSI Read/Write Frame Count and SCSI Read/Write Data statistics are supported only for
F_Ports for any flow configuration where either the source device or destination device exists
on the switch, and the flow is defined using a combination of source device, destination
device, ingress port, or egress port (with or without bidirectional), or a combination of source
device, destination device, LUN, ingress port, or egress port.
•
Integration with the Monitoring and Alerting Policy Suite (MAPS) service to enable
threshold-based monitoring and alerting based on flows. For more information on integration
with MAPS, refer to the
“Monitoring and Alerting Policy Suite integration with Flow Vision” on
page 1214".