Brocade Virtual ADX Server Load Balancing Guide (Supporting ADX v03.1.00) User Manual
Page 42
26
Brocade Virtual ADX Server Load Balancing Guide
53-1003247-01
Changing the Load-Balancing Predictor Method
2
To smooth the samples out, the Brocade Virtual ADX uses the following calculation:
R = ((S * previous_R) + ((100 - S) * new_R)) / 100
where:
R = Response time
S = smooth factor, which is configurable and can be from 1to 99. The default is 60. A large
value gives the previous response time more weight than the new response time. A small value
gives the new response time more weight than the previous response time.
For example, if a given real server’s previous response time value was two milliseconds, and a new
measurement also results in two milliseconds, the calculated server response time using the
smooth factor of 90 is as follows:
R = ((90 * 2) + ((100 - 90) * 2) / 100
R = 180 + 20 / 100
R = 200 / 100
R = 2
If the real server’s response time slows due to processing for additional connections, the slower
response time affects the Server Response Time calculation for the server. For example, if the next
server response time sample is five milliseconds instead of two, the Server Response Time
calculation is as follows:
R = ((90 * 2) + ((100 - 90) * 5) / 100
R = 180 + 50 / 100
R = 230 / 100
R = 2.3
Since the real server’s response time has slowed by two and a half times, the server’s response
time calculation biases the Brocade Virtual ADX away from selecting that server for new
connections.
You can affect the degree of difference in subsequent response time weights by changing the
smooth factor (S). For example, if you change the smooth factor from 90 to 50, the calculations
shown above have the following results:
Here is the calculation when the previous and new response times are 2 milliseconds:
R = ((50 * 2) + ((100 - 50) * 2) / 100
R = 100 + 100 / 100
R = 200 / 100
R = 2
Here is the calculation if the server’s next response time is 5 milliseconds.
R = ((50 * 2) + ((100 - 50) * 5) / 100
R = 100 + 250 / 100
R = 350 / 100
R = 3.5