Brocade Virtual ADX Server Load Balancing Guide (Supporting ADX v03.1.00) - Manual (Page 25)

Brocade Virtual ADX Server Load Balancing Guide 9
53-1003247-01
Overview
2
Weighted Round Robin predictor
Like the Round Robin predictor, the Weighted Round Robin predictor treats all servers equally
regardless of the number of connections or response time. It does however use a configured
weight value that determines the number of times within a sequence that the each server is
selected in relationship to the weighted values of other servers. For example, in a simple
configuration with two servers where the first server has a weight of 4 and the second server has a
weight of 2, the sequence of selection would occur as described in the following:
1. The first request is sent to Server1.
2. The second request is sent to Server2.
3. The third request is sent to Server1.
4. The fourth request is sent to Server2.
5. The fifth request is sent to Server1.
6. The sixth request is sent to Server1.
Notice that over this cycle of server connections, Server1, which had a weight of 4, was accessed
four times and Server2, which had a weight of 2, was accessed only twice.
This cycle will repeat as long as this predictor is in use.
The Weighted Round Robin predictor can be applied globally to the entire Brocade Virtual ADX or
locally per virtual server as described in
“Changing the Load-Balancing Predictor Method” on
page 21.
Static Weighted Round Robin predictor
The Static Weighted Round Robin predictor makes its server selections exactly like the Weighted
Round Robin predictor, however, it does not distribute the load to available barrel processors (BPs)
within the Brocade Virtual ADX. Consequently, server selection can be concurrent to better utilize
your system capacity. The following description provides a simple example:
The Brocade Virtual ADX has the following configuration:
• Two BPs are enabled and in an operating state
• Three servers are connected: Server1 with a weight of 3, Server2 with a weight of 2, Server3
with a weight of 1
Distribution will occur as described in the following.
For BP1
1. The first request is sent to Server1.
2. The second request is sent to Server2.
3. The third request is sent to Server3.
4. The fourth request is sent to Server1.
5. The fifth request is sent to Server2.
6. The sixth request is sent to Server1.