Do you work in an environment where a Teradata compression tool is installed that conforms to your satisfaction?
Part of the Teradata Compression and Space Management guide.
If not, I assume you are probably not very pleased with having to walk through the databases, table by table, to find out what to compress on which column.
Typically, this is time-consuming, fatiguing, and subject to momentary human judgment.
Every reevaluation is costly again. Therefore, Teradata compression activities often stop once the largest decile of tables has been treated, leaving a lot of potential unused. I can offer you a solution to this problem:
What Is The Teradata DWHPRO TITHONIZER?
TITHONIZER selects compression candidates using a greedy savings-to-header-cost heuristic. This can find useful savings, but does not establish a globally optimal combination under every constraint.

DWHPRO TITHONIZER determines a set of compression candidate columns and the compression candidate values per column.
It can be used for compressing a table for the first time or an update on given compressions.
It does not apply the compression findings directly. Instead, it stores them in a separate table to be implemented at a later stage.
This allows you to separate the potentially resource-intense analysis from the application work.
Get the complete Teradata tuning book, free
Teradata Query Performance Tuning: The DWHPro Guide
By Roland Wenzlofsky & Artemios Vogiatzis
206 pages on statistics and demographics, access paths and indexes, query execution plans, and joins. Get the full PDF and practical Teradata updates from DWHPro.
Confirm your email to receive your download link. Privacy policy. Unsubscribe in any newsletter.
About the free bookAlso, DWHPRO TITHONIZER can be applied to one table in isolation or all of a database’s tables. This enables you to perform demo or study runs over representative tables before any mass activity.
It gathers the following information:
- Is the table populated with enough compression rows to have a savings potential?
- Which columns can be candidates for compression for technical reasons?
- Which columns shall be candidates for compression from a business or policy point of view?
- Which values, given columns, are candidates?
- What is the benefit of saved bytes of compressing one value?
- What is the cost of compressing this one value?
- What is the Presence Byte consumption of a given compression in total?
Compressed value sets for every column are summarized in levels representing one Presence Bit consumed each.
Based on this information, a dominant path for compression is determined by ordering the various levels of columns by the “savings per header cost ratio”.
Highly frequent level 1 values with little cost in bytes for the table header come first in line. Conversely, one will find lengthy and infrequent values with razor-thin or no net benefit.
After potentially adjusting for the net benefit and local Presence Byte break-even points, the dominant path is translated into a series of compression statements stored in the result table TITHONOS. From there, alter table statements can be generated to apply the compression.
For a detailed description of DWHPRO TITHONIZER, please refer to the pdf file attached to this article:
I recommend reading it before you start working with the Stored Procedure.
DWHPRO TITHONIZER was developed and tested for Teradata Versions 13 and 14, 15, and 16.
Improvements, refinements, or modifications of functionality or methodology will occur as experience with it is gained and time to do it is available.
I also hope for valuable input from you as our readers. If you need additional features or find any bugs, send us an email:
Feel free to use or modify DWHPRO TITHONIZER for training or working purposes. But please note in the program code that the original version is from DWHPRO.
Before you start finding out what the Stored Procedure can do for you, let me remind you that the responsibility for using it and any consequence thereof lies wholly in the hand of whoever uses it.
I am not responsible for any damage, misuse, or other negative consequences from DWHPRO TITHONIZER applications.
If you have gathered experience, I would be pleased to hear of any results you achieved with it:
Download the original historical TITHONIZER source
On Teradata Database 20.00.28.81, the original procedure required seven hexadecimal newline literals to be replaced with CHR(10) for UTF-8 compilation. Use the corrected source and the documented G00_SP_LOG_TEMP logging table, compile-time volatile-table cleanup and scoped owner SELECT rights. The adapted workflow ran on 10,000 synthetic rows and generated candidates; the historical storage saving was not reproduced.
Download the corrected UTF-8 TITHONIZER source and setup notes. This version replaces seven hexadecimal newline expressions with CHR(10). It compiled and ran on the documented synthetic fixture; follow the original installation prerequisites and review the generated SQL before use. It does not establish globally optimal compression or reproduce historical storage savings.
Need help with query performance or platform costs?
Diagnose plans, skew, spool pressure and expensive workloads. Agree a baseline and acceptance criteria before making changes.
Get the complete Teradata tuning book, free
Teradata Query Performance Tuning: The DWHPro Guide
By Roland Wenzlofsky & Artemios Vogiatzis
206 pages on statistics and demographics, access paths and indexes, query execution plans, and joins. Get the full PDF and practical Teradata updates from DWHPro.
Confirm your email to receive your download link. Privacy policy. Unsubscribe in any newsletter.
About the free book
Hi,
I am awaiting for the tool link , where i can download and start trying it. I have also sent a message to [email protected]. Can you please help me here.
HI. Sorry, for the delay. I will put the latest version online at the begin of next week.
Please send me the steps and download link for the compresion. I also sent an email from [email protected]
required for applying complression
Hi Roland
Have you had a chance to upload the latest version yet, can’t see it?
Cheers
Steve
Hi
I thought I posted something last week, but nothing there, so here goes again.
I’m trying to get the stored procedure to run,
CALL DWHPRO_TITHONIZER(‘DWHProCompression’, ”);
but keep getting the error:
Executed as Single statement. Failed [3707: 42000] DWHPRO_TITHONIZER: Syntax error, expected something like ‘)’ or an ‘EXCEPT’ keyword or a ‘UNION’ keyword or a ‘MINUS’ keyword between a string or a Unicode character literal and the word ‘DWHProCompression’.
Elapsed time = 00:00:09.503
STATEMENT 1: CALL failed.
I’ve tried running through Studio and SQLA, get the same error.
We are running TD 15.00
Cheers
Steven
HI. I will uploade the latest version later today. Maybe this fixes the error.