Teradata Workload Analyzer enables DBAs to save time and improve system management.


Applied Solutions

Drill Down

Teradata Workload Analyzer enables DBAs to save time and improve system management.

DBAs often spend a lot of time setting up workloads, which can involve a user-intensive trial-and-error process. But there’s an easier way: Teradata Workload Analyzer. This solution helps DBAs fully understand normal work­loads, peak surges and abnormal workloads in multiple time periods so they can design the correct workload definitions. This leads to higher throughput and more satisfied business users, and it ensures that users only add more server capacity when they need it. In short, Teradata Workload Analyzer:

  • Improves workload accuracy and quality
  • Enables improved response time consistency
  • Increases predictability of workloads
  • Enhances system management capabilities
Figure: Workload Cluster Flowchart

Click to enlarge

Drill for Answers

Teradata Workload Analyzer provides the capabilities to continuously drill down on a workload with various “Who,” “What,” “Where” and “Exceptions” parameters. It can visualize clusters of requests within the workload, each with distinct service time patterns and other characteristics. For example, an initial workload defined on just one account may include distinct users who execute tactical requests requiring higher priority, while the remaining users do not. Workload Analyzer can be used to identify these clusters by providing reports, correlation and distribution graphs on dif­ferent dimensions.

Workload Analyzer guides DBAs toward appropriate classification criteria. (See figure.) At any point, a visual drill down workload cluster analysis can be performed where correlation and distribu­tion parameters can be chosen and then analyzed for associated usage patterns. Drilling deeper within a chosen cluster can determine which parameters most effec­tively identify the request group to isolate. This trial-and-error approach streamlines the process by providing users with distinct count and distribution range insight and can eliminate ineffective analysis on a single user or a tight distribution.

Set New Parameters

Drilling down is a recursive process for deeper analysis on correlation and distribu­tion parameters. If DBAs are not satisfied with the current analysis parameters, they can select more appropriate ones, as guided by reviewing distinct values and ranges for those other parameters. For example, with respect to the distinct value counts, one particular workload could display:

  • User Name (24)
  • Applications (1)
  • Account Name (1)
  • Client Addresses (2)
  • Queryband (3)
    • Function (3)
    • Urgency (1)
    • AggLevel (8)
  • Estimated Processing Time (0-1000 secs)
  • AMP Count (0-1)

Improved Capabilities, Reduced Effort

Teradata Active System Management is a portfolio of products designed to manage data warehouse appliances and integrated data warehouses. It greatly enhances Teradata work­load management capabilities and reduces the effort required by DBAs, application developers and support engineers to man­age workloads.

Teradata Workload Analyzer, a component of Teradata Active System Management, helps optimize any data ware­house. It collects system usage information from database logs and system tables, analyzes the historical system use, and delivers reports and graphical results. With this information, DBAs can produce high-quality workload definitions that con­trol the system resources.

In this instance, DBAs find that only one distinct application or account exists and all queries are run at the same urgency. With this information, trying to identify a corre­lation against different Application, Account or Urgency values would be a wasted effort.

However, the opportunity for correla­tion does exist with Users, Function and AggLevel, so those correlation options can be pursued. Similarly, for the distribution parameter ranges, an estimated processing time range from 0 to 1,000 seconds sug­gests that a large variation of requests is included in this workload. The opportu­nity for identifying clusters is higher than if the estimated processing time range were simply 0 to 1 second.

DBAs can add clusters to the current workload for deeper analysis, or clusters can be split into a new workload. By repeating this process, a good set of workloads can be defined or all unassigned clusters can be assigned to workloads.

DBA Time Saver

Workload Analyzer will reduce DBAs’ efforts to set up Teradata Active System Management workloads and increase their quality with fewer correlation attempts. Improved response time consistency and predictability of workloads will also be realized as well as improvements in system management capabilities.

Workload Analyzer provides a better solu­tion than other methods by establishing and refining workloads while cutting the time and effort spent on unsuccessful attempts.

Your Comment:
Your Rating:

Fuzzy Logix