Skip to Content
Query Performance, Pareto, Goldratt’s TOC & Me  BIM 205 at TechED’06 Bangalore presented A Quantitative Approach to query performance that talks about following steps for query improvement –   1.Measure: What is current query performance? What are different parameters to look at? What are the tools to get these parameters? What parameters need to be looked into? How are these parameters measured?  2.Analyze: What do you make of current performance parameters?  3.Action: Which corrective actions are relevant based on Analyze phase?   4.Improvement: How to measure the improvement during the post action phase?  5.Control: How do you deliver consistent optimized performance over longer period of time?   That was the Approach, here is how could you couple it with other leading management concepts and apply it effectively. Quantitative Approach can be coupled with world famous concepts of “Pareto Principle” and Goldratt’s “Theory of Constraints”.   To start with what are these concepts –  1.Pareto’s Principle The 80/20 rule that is based on Pareto’s research that states only a vital few, (20%), of causes will have a greater impact than the many, (80%), causes.  Examples: 80% of the problems come from 20% of the causes 80% of a company’s defects are caused by from 20% of products 2.Theory of Constraints by Goldratt    a. Identify the constraint    b. Exploit it    c. Subordinate all other operations to the necessity        to  exploit the constraints    d. Elevate the constraint post steps “b” & “c”    e. Go back to step “a”, but don’t let inertia become        the system constraint    When applying the Quantitative Approach for Query Performance I suggest starting with the Pareto principle and concentrate on –  1. Most Frequently used Queries ( must be around 20% of total no of queries) 2.Out of these most frequently used ones the top 20% consuming maximum time or the critical ones 3.So no of queries you need to concentrate in case you have total 100 queries are 100*.20*.20 = 4 And during the Control phase of Quantitative Approach consider “Theory of constraints” i.e. every system has a constraint. Any effort in improving anything other than ‘the’ constraint is wasteful to the system in that it adds no value to the overall throughput. When you relieve (open up) one constraint, then bottleneck moves to the next tightest one [From Goldratt Theory of Constraints].  So as soon as you are done with optimization of these “4” queries another set of “4” queries will raise their head to be optimized. That is why quantitative approach mentions on Query Performance Improvement – “It is a continuous Improvement process to consistently deliver optimized performance” 
To report this post you need to login first.

3 Comments

You must be Logged on to comment or reply to a post.

  1. Gopi S K
    Thanks to Vikash.
    You have provided a good approach to improve the performance more towards a quality management prospective.
    How far can we link this to technical process? I mean, how can we map technical challenges to this?

    Regards
    GSK

    (0) 
    1. Vikash Agrawal Post author
      Hi GSK,

      Thanks.

      you can go through the TechEd Material of BIM205 (Bangalore), there I have tried to cover the technical aspects of it.

      Regards

      Vikash

      (0) 

Leave a Reply