Wednesday, February 03, 2010

January 2010 Uptime Reports for OakLeaf Azure Test Harness Running in the South Central US Data Center

Update 2/3/2010: Following is a screen capture the Pingdom monthly report for 1/1 through 1/31/2010, which arrived Wednesday morning:

Pingdom01-2010-547px

Pingdom reported substantially more downtime (110 minutes) than mon.itor.us (9 minutes). Following is a screen capture of the detailed downtime report for the period (Azure Tables(1).csv):

Pingdom01-2010Details838px 

Following are mon.itor.us Uptime reports for the the OakLeaf OakLeaf Systems Azure Table Services Sample Project from Cloud Computing with the Windows Azure Platform running in Microsoft’s San Antonio, Texas (South Central US) data center for the first four weeks of Windows Azure’s commercial operation:

MonitorUsJanuary2010-974px 

Total reported downtime for the month was 0.15 hours = 9 minutes for a monthly uptime of 99.93%. Only a single instance of oakleaf.cloudapp.net is running, so the Microsoft service level agreement (SLA) guarantee of 99.9% uptime doesn’t apply. However, the results were well within the SLA limits.

My MSDN Premium subscription benefits subsidize only one compute instance, so the other two test harnesses (AzureBlob: oakleaf2.cloudapp.net and AzureQueue: oakleaf5.cloudapp.net) shown in December 2009 Uptime Reports for OakLeaf Azure Demos Running in the South Central US Data Center no longer appear in the results. These two instances were removed on 1/6/2010, so the Total Average Resp(onse) Time and OK values for Week N1 don’t compute.

blog comments powered by Disqus