Tuning for Peak Load

  • Ben DeBow


Today is historically the busiest day of the year for your online shopping business. At 11 AM, the system is starting to show signs of strain and load that exceeds that of the previous year. Customers are having problems completing their orders. Some fail, others are just slow. The operations center calls you in to see what’s going on with SQL Server, and you observe that processes are starting to queue up because of large blocking chains, which ultimately affects the checkout process. You didn’t code the application, but your job as a database administrator (DBA) is to fix the problem. As this event is unfolding, upper management is concerned and the Chief Information Officer (CIO) is standing over your shoulder watching you try to resolve the issue—no pressure there. Things eventually calm down later in the day, but it is unknown how many sales or potential customers were lost because of the slowdown or failed checkouts. In addition to the orders being affected, replication to the reporting environment now has 12 hours of latency affecting the reports to internal consumers of the raw sales data. Can you prevent this scenario from happening? In many cases, you can.


Peak Load Network Inter Face Card Chief Information Officer Performance Monitor Object Counter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Bradley Ball, TJay Belt, Glenn Berry, Jes Borland, Carlos Bossy, Louis Davidson, Jeremy Lowell, Ben DeBow, Grant Fritchey, Wendy Pastrick, Kellyn Pot’vin, Jonathan Gardner, Jesper Johansen, Mladen Prajdić, Herve Roggero, Chris Shaw, Gail Shaw, Jason Strate 2012

Authors and Affiliations

  • Ben DeBow

There are no affiliations available

Personalised recommendations