Database Performance

  • Peter Lake
  • Paul Crowther
Part of the Undergraduate Topics in Computer Science book series (UTICS)

Abstract

For a DBA the dreaded phone call is “my report is running very slowly”. For a start, what is mean by slowly? What is the user used to? Then there is the problem of how you establish where the problem is—is it hardware related? Or Network related? At the Server or Client end? We see that performance problems can be hard to diagnose because of the complexity of the tiers involved (physical, OS, network, servers, etc). Moreover, even if we do find the cause and resolve the problem, there is no guarantee that it will remain solved because, as we add data to our system, the data distribution may make some of our previous design assumptions invalid. The solution may be indexes, or partitions, or a variety of other performance enhancing techniques that we review in this chapter. We include some tutorial material which explores some performance management tools. We look at some of the important elements of a databases performance, discussing issues like throughput, workload and contention. We note that there are many levels at which we can apply performance tools, including the physical level (increasing RAM for example), and at the Schema level. However, much of our focus is on SQL Optimisation, and we provide examples of using tools like Explain Plan to assist with the tuning of SQL.

Keywords

Expense Sorting Rounded 

References

  1. Florescu D, Kossmann D (2009) Rethinking cost and performance of database systems. SIGMOD Rec 38(1):43–48. 2009 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  • Peter Lake
    • 1
  • Paul Crowther
    • 1
  1. 1.Sheffield Hallam UniversitySheffieldUK

Personalised recommendations