SharePoint world of ECM and Information Management

SharePoint Tip #17. Do you know “why to measure network latency of SharePoint Farm”?

Network response time is one of the important factors that affect SharePoint farm design. Measure the latency between SharePoint servers and users to reorganize servers according the smallest response time. This factor is a key point consideration to determine which of the supported deployment solutions to implement for your Office SharePoint Server deployment. (Latency is the time required for a packet to travel from one point on a network to another).

Use the Ping tool (ping.exe) to measure latency, which affects:

  • users - from the client computer to the Web server on the server farm
  • data centres, that host servers of the same farm - from a Web server in the remote data centre to the database server in the primary data centre

Do not forget to divide the round-trip result by two, because all measures are one way only, not round-trip. Compare results to the data below, and adopt environment to have latency lower those values. 

Number of users  

Concurrent users (10%)  

Central Solution  

Distributed solution  

100-5,000  

10-500  

Bandwidth:   3+ Mbps (dual T1)  

Latency:   < 100 ms  

Bandwidth:   1.5 Mbps (T1)  

Latency:   <100 ms  

10,000  

1,000  

Bandwidth:     3+ Mbps (dual T1)  

Latency:   <250 ms  

Bandwidth:   1.5 Mbps (T1)  

Latency:   <500 ms  

100,000  

10,000  

Bandwidth:   3+ Mbps (dual T1)  

Latency:   < 250 ms  

Bandwidth:   1.5 Mbps (T1)  

Latency:   <500 ms  

Overstepping these values will increase the page-load times dramatically, in 4 times at least. The critical bandwidth is 1.5 Mbps (T1) with 500ms latency.

 

Available network bandwidth and existed latency influences planning for geographic deployments in which data travels across WAN links that span multiple cities, states, provinces, countries, or continents.

 

Have anything to add?! Send your tips to be published via this form

Comments

Paul Grenier said:

When testing with ping, the default packet sizes are very small.  By watching typical traffic, you can shape test packets to the high, low, and average sizes.  This can help alert designers to problem areas before they appear in production.

# February 25, 2009 7:50 AM
Leave a Comment

(required) 

(required) 

(optional)

(required)