Teradata Query Execution Logs Extraction and Source Code Prerequisites

This topic briefs about the Teradata assessment query execution logs extraction and source code prerequisites.

In This Topic:

Introduction
Assessment Process
Source Code Assessment

Introduction

LeapLogic Assessment profiles existing inventory, identify complexity, lineage and provides comprehensive recommendations for migration to modern data platform.

Assessment checklist

Information	Format	Availability?
Teradata Query Logs	File Export in text format	Yes/ No
Total I/O Usage by Days	File Export in text format	Yes/ No
Teradata Hardware Config for Total CPU Calculation	Inline in the document	Yes/ No
Database Object Count	File Export in text format	Yes/ No
Database Volume	File Export in text format	Yes/ No
High Data Volume Tables	File Export in text format	Yes/ No
Total Number of Databases and Users	File Export in text format	Yes/ No
Data for Partitioning bucketing	File Export in text format	Yes/ No
Source Code	File Export	Yes/ No

Follow the steps in section 2, 3 to collect the required information for assessment.

Assessment Process

All the environment specific variables are highlighted. The start and end date should be consistent for all the queries.

Logs

This requires query execution logs in a CSV format. The CSV file must have the following columns (~~ separated file) retrieved from Dbqlogtbl and QryLogSQL tables of PDCRInfo/DBC database.

FILLER~~UserName~~AppID~~ClientID~~StartTime~~AMPCPUTime~~TotalIOCount~~ParserCPUTime~~FirstRespTime~~FirstStepTime~~ProcID~~QueryID~~MaxAMPCPUTime~~MaxAmpIO~~totalCPU~~totalIO~~Query_Execution_Time~~OBJECTDATABASENAME~~SQLROWNO~~StatementType~~ LogonSource~~ QUERYBAND~~SQLTEXTINFO

Two options are available for extracting query execution logs: the first involves utilizing the TPT script, while the second involves using FastExport. The recommended approach is to use the TPT script for query log extraction.

Option 1: TPT Export

An effective approach to extract information from query logs is by utilizing a TPT script. This method uses the TPT utility to retrieve the query execution logs, resulting in the generation of a log specific to the provided start and end dates.

The typical steps are:

Copy the following artifact into a folder on the system where TTU (Teradata Tools and Utilities) is installed.

QueryLogExtract Download

Next, modify the specified properties in the file.
- Update the destination path for storing the query log output.

VARCHAR DirectoryPath=‘/home/ec2-user/output’,

update the TPID with the real Teradata system name or IP address.

VARCHAR Tdpid=‘54.234.225.211’,

Update the Teradata system’s username and password with the real credentials.

VARCHAR UserName=’dbc‘, VARCHAR UserPAssword=’dbc‘,

Lastly, execute the TPT script using the provided command, ensuring to adjust it accordingly to specify the desired start and end dates for the query execution log.

tbuild -f /path/QueryLogExtract.tpt -u “start_date=‘2023-07-01’, end_date=‘2023-07-20’“| tee QueryLogExtract.tpt.log

Option 2: FastExport

An alternative approach to extract information from DBQL logs involves using FastExport. However, it’s important to keep in mind that utilizing FastExport may result in larger output file size.

The typical steps are:

Save the below fast export script with a filename ‘TD_fastexport.fe’

.LOGTABLE dbname.fastexp_log;

.LOGON TDPID/username,password;

DATABASE dbname;

.BEGIN EXPORT SESSIONS 10;

.EXPORT OUTFILE D:\path\Teradata_fastExp_query_log.txt

MODE RECORD FORMAT TEXT;

select cast(‘IDWWM’||’~~’||

rtrim(cast(tb1.Username as char(128)))||’~~’||

rtrim(cast(tb1.AppID as char(30)))||’~~’||

rtrim(cast(tb1.ClientID as char(30)))||’~~’||

rtrim(cast(tb1.StartTime as char(21)))||’~~’||

rtrim(cast(tb1.ampcputime as char(10)))||’~~’||

rtrim(cast(tb1.TotalIOCount as char(20)))||’~~’||

rtrim(cast(tb1.ParserCPUTime as char(10)))||’~~’||

rtrim(cast(tb1.firstresptime as char(21)))||’~~’||

rtrim(cast(tb1.firststeptime as char(21)))||’~~’||

rtrim(cast(tb1.ProcID as char(6)))||’~~’||

rtrim(cast(tb1.QueryID as char(20)))||’~~’||

rtrim(cast(tb1.MaxAMPCPUTime as char(10)))||’~~’||

rtrim(cast(tb1.MaxAMPIO as char(20)))||’~~’||

rtrim(cast(tb4.ampcputime as char(10)))||’~~’||

rtrim(cast(tb4.TotalIOCount as char(20)))||’~~’||

rtrim(cast(((CAST((CAST(tb1.firstresptime AS DATE)- CAST(tb1.firststeptime AS DATE)) AS DECIMAL(18,6)) * 60*60*24)

+ ((EXTRACT( HOUR FROM tb1.firstresptime) – EXTRACT( HOUR FROM tb1.firststeptime)) * 60*60)

+ ((EXTRACT(MINUTE FROM tb1.firstresptime) – EXTRACT(MINUTE FROM tb1.firststeptime)) * 60)

+ (EXTRACT(SECOND FROM tb1.firstresptime) – EXTRACT(SECOND FROM tb1.firststeptime)) ) as char(20)))||’~~’||

rtrim(cast(coalesce(TB3.OBJECTDATABASENAME,’Not Present’) as char(500)))||’~~’||

rtrim(cast(tb2.sqlrowno as char(2)))||’~~’||

rtrim(cast(tb1.statementtype as char(100)))||’~~’||

rtrim(cast(tb1.logonsource as char(100)))||’~~’||

rtrim(cast(coalesce(tb1.queryband,’Not Present’) as char(6160)))||’~~’||

rtrim(cast(tb2.SQLTEXTINFO as char(20000)))

as char(27000))

)

from

pdcrinfo.dbqlogtbl tb1 inner join pdcrinfo.dbqlsqltbl tb2

on tb1.queryid = tb2.queryid

and tb1.procid = tb2.procid

and tb1.logdate = tb2.logdate

left outer join

(

select queryid,procid,logdate,tdstats.udfconcat(trim(objectdatabasename)) as objectdatabasename

from pdcrinfo.dbqlobjtbl

where objecttype=’db’ and logdate between ‘2017/01/01’ AND ‘2017/05/29’

group by queryid,procid,logdate

)

tb3

on tb1.queryid = tb3.queryid

and tb1.procid = tb3.procid

and tb1.logdate = tb3.logdate

inner join

(

select logdate,sum(ampcputime) ampcputime,sum(totaliocount) as totaliocount

from pdcrinfo.dbqlogtbl

where

logdate between ‘2023/01/01’ AND ‘2023/05/29’

group by logdate

)tb4 on tb4.logdate = tb1.logdate

where

tb1.logdate between ‘2023/01/01’ AND ‘2023/05/29’

and tb1.errorcode=0

order by tb1.procid,tb1.queryid,tb2.sqlrowno;

.END EXPORT;

.LOGOFF;

Highlighted variable to be modified as:
- Replace ‘2023/01/01’ AND ‘2023/05/29‘ with actual export assessment start and end date within the attached query.
- Replace dbname with actual database name.
- Replace ‘TDPID’ with actual Teradata Server IP.
- Replace ‘username, password’ with actual database username and password within the attached query.
- Replace database name ‘dbc’ with actual database within the attached query.
- Replace DATAFILE path ‘D:\path\Teradata_fastExp_query_log.txt’ with actual path and fileName within the attached query.
Execute the FastExport script.

fexp < TD_fastexport.fe

Notes for Troubleshooting:

If the user does not have access to function tdstats.udfconcat then:

Grant the required permission to user before running the script again

GRANT SELECT ON TDSTATS to “USERNAME”;

Change “tdstats.udfconcat(trim(OBJECTDATABASENAME)) as OBJECTDATABASENAME” to “trim(OBJECTDATABASENAME) as OBJECTDATABASENAME”
Grant CREATE TABLE and DROP TABLE privileges to the Teradata user to create the fast export log table

GRANT CREATE TABLE, DROP TABLE

ON dbname

TO userid;

If the TPT script is modified on Windows machine and uploaded to Linux system, then,

use dos2unix tool to convert text files from DOS line endings (carriage return + line feed) to Unix line endings (line feed).

# dos2unix QueryLogExtract.tpt dos2unix: converting file QueryLogExtract.tpt to UNIX format …

Total I/O Usage by Days

Save the results of this query in a CSV file:

SELECT THEDATE AS RUNDATE,

CAST(SUM(FILEACQREADS+FILEPREREADS+FILEWRITES) AS BIGINT) AS TOTALIOREADS

FROM DBC.RESUSAGESPMA

WHERE THEDATE BETWEEN ‘2023/01/01’ AND ‘2023/05/29’ GROUP BY THEDATE;

Replace ‘2023/01/01’ AND ‘2023/05/29’ with actual export assessment start and end date.

Hardware Configuration

The details of hardware configuration are required for the calculation of total CPU utilization. Please provide the following details.

Teradata series and specs (Cores, RAM, HDD/SDD)
Number of Teradata nodes
TPerf
Total available AMPCPU seconds per day

Database Objects

This query provides the count of database objects in use. Save its results in a CSV file.

select databasename,tablekind,count(tablename) from dbc.tables group by databasename,tablekind order by databasename

Database Volume

This SQL will collect databases with volume equal or above 10GB. Save its results in a CSV file.

select databasename,cast (sum(currentperm)/1024/1024/1024 as decimal(18,2))

from dbc.diskspace

group by databasename having cast (sum(currentperm)/1024/1024/1024 as decimal (18,2)) >= 10 order by 2 desc

Note:

The filter condition can be changed for collecting lower volumes.

High Data Volume Tables

This query provides result for tables with high data volume. Save its results in a CSV file.

select

ts.databasename,

ts.tablename,

rc.num_rows,

ts.currentsize_gb,

pv.ConstraintType,

pv.ConstraintText

from

(select

databasename,tablename,sum(currentperm)/1024/1024/1024 as currentsize_gb

from

dbc.tablesize

group by 1,2

having sum(currentperm)/1024/1024/1024 >= 10

)ts

left join

(select databasename,tablename,max(sv.rowcount) num_rows from dbc.statsv sv group by databasename,tablename) rc

on ts.databasename=rc.databasename and ts.tablename=rc.tablename

left join DBC.PartitioningConstraintsv pv

on ts.databasename=pv.databasename and ts.tablename=pv.tablename order by ts.currentsize_gb desc

Note:

The filter condition can be changed for collecting lower volumes.

Databases and Users

This query provides the total number of databases and users. Save its results in a CSV file.

select dbkind,count(*) from dbc.databases group by dbkind

Data for Partitioning / Bucketing

This query extracts table size and column details to be utilized for partitioning and bucketing recommendations. Save the results in a CSV file.

SELECT cols.databasename AS DATABASE_NAME,

cols.tablename AS Table_Name,

Cast(alsp.currentperm AS BIGINT) AS TABLE_SIZE_BYTES,

” AS NUM_ROWS,

cols.columnname AS Column_Name,

” AS num_unique_val

FROM dbc.columns AS cols

JOIN ( select databasename,tablename, cast(sum (currentperm) as bigint) as currentperm from dbc.allspace group by 1 ,2 ) alsp

ON cols.tablename = alsp.tablename and cols.databasename = alsp.databasename

WHERE cols.tablename <> ‘All’;

Source Code Assessment

Provide the below active or in-scope Teradata source code artifacts as applicable in the migration scope.

Code Artifact	Criticality	Remarks
Orchestration Scripts (Control-M / Autosys / Cron etc.)	Must	To identify interdependencies across scheduler scripts / jobs, queries, and dependent workloads
Procedures / Functions	Must	To identify workload complexity, query patterns, query count etc. for effort estimation and technical debt
Views	Must	To identify view complexity, patterns and effort estimations
Shell Scripts	Must	To identify count, dependencies, SQL queries and PL/SQL, logic (example: email, notification etc.) and effort estimations
DDL	Must	To identify column usage, and provide recommendation on column level lineage, and query optimization on the target system
DML / SQL Files	Must	To identify count, dependencies, SQL queries and effort estimations
CTL Files	Must	To identify source data type and formats, example Delimiters, JSON etc.
BTEQ	Must	To identify count, dependencies, SQL queries and PL/SQL, transformation logics and effort estimations
TPT, Fexp, mLoad	Must	To identify count, complexity of data ingestion scripts and effort estimation

Note:

Limit: Assuming the orchestration script is a trigger point for every single use case execution in the existing setup. If customer is not comfortable sharing all the workloads, then share those workloads which are referred or executed through the orchestration scripts. However, in such scenarios the scope and effort estimates will be based on the given workloads.