Browse Prior Art Database

A method for a performance-predictive planning UI for placement of database tasks used by database administrators

IP.com Disclosure Number: IPCOM000201998D
Publication Date: 2010-Nov-30
Document File: 5 page(s) / 88K

Publishing Venue

The IP.com Prior Art Database

Abstract

Today, database administrators (DBAs) struggle with non-integrated toolsets for accessing performance history data and performance monitoring, database administration and development and task planning. The purpose of this article is to describe a novel method integrating key information parameters into a performance-predictive planning UI for database administrators. Essential elements include showing performance relevant data in the context of a task planning UI with automatically suggesting the best slot(s) to place a new administrative task such as the next backup or reorg task.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 22% of the total text.

Page 01 of 5

A method for a performance-predictive planning UI for placement of database tasks used by database administrators

As of today, there are 3 non-integrated tool sets:
database performance monitoring/performance advisory tools for improving database performance
database administration/development tools
planning /task management tools

The typical database administrator (DBA) has on a daily basis the challenge to schedule workloads and administrative tasks for the database with no overall insight what is running at the same time. Tools such as DB2 Performance Expert can provide insight how the resource consumption looked like in the past (memory, IO, CPU) or provide insight how much resources might be required by a SQL Query/workload in the future (using EXPLAIN capabilities of the administrative/development tools). However, this information is not presented in a coherent fashion alongside planned activities in a planning tool. Furthermore, prediction capabilities about system utilization regarding IO/memory/CPU based on average consumption in the past is not visualized alongside a planning / task management tools so that the DBA can easily determine when it would be a good time to schedule additional workloads and administrative tasks. This means, the DBA needs to spend significant time researching and aligning all the information on resource consumption/predicted resource consumption across a heterogeneous tool set using multiple different UIs which is time consuming and error-prone which creates the following issues:

Lack of context-sensitive presentation of workload history in the planning UI used by the DBA
Lack of conflict detection for logically conflicting workloads
Lack of showing relevant, extraordinary days in the planning UI such year-end processing of data, predicted day the database runs out of storage if no additional, physical storage is added, public holidays which would allow additional administrative workload on the system, etc.

For any scheduled workload/event in the database, there might be a history related to it the DBA should know (e.g. previous schedule of workload failed due to xyz, was running too slow due to missing performance tuning of SQL, etc.)

In the planning UI there is no indication which are permissible windows for offline/online maintenance work - without that information its hard to determine for the DBA if a maintenance task fits into the maintenance window
Lack of context-sensitive presentation of performance data for the relevant time window (e.g. if a week is shown, based on previous performance measurements and prediction of performance characteristics of planned tasks, all the relevant metrics are shown next to the week shown, if a month is shown it automatically shows it for the month, etc.

Lack of context-sensitive presentation of estimated resource consumption of the planned task and impact on the system based on the selected - but not yet committed - location of the task. This includes estimat...