PI World 2020 Lab

[Pages:39]PI World 2020 Lab Optimizing Analytics for Better Performance

OSIsoft, LLC 1600 Alvarado Street San Leandro, CA 94577 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, mechanical, photocopying, recording, or otherwise, without the prior written permission of OSIsoft, LLC. OSIsoft, the OSIsoft logo and logotype, Managed PI, OSIsoft Advanced Services, OSIsoft Cloud Services, OSIsoft Connected Services, OSIsoft EDS, PI ACE, PI Advanced Computing Engine, PI AF SDK, PI API, PI Asset Framework, PI Audit Viewer, PI Builder, PI Cloud Connect, PI Connectors, PI Data Archive, PI DataLink, PI DataLink Server, PI Developers Club, PI Integrator for Business Analytics, PI Interfaces, PI JDBC Driver, PI Manual Logger, PI Notifications, PI ODBC Driver, PI OLEDB Enterprise, PI OLEDB Provider, PI OPC DA Server, PI OPC HDA Server, PI ProcessBook, PI SDK, PI Server, PI Square, PI System, PI System Access, PI Vision, PI Visualization Suite, PI Web API, PI WebParts, PI Web Services, RLINK and RtReports are all trademarks of OSIsoft, LLC. All other trademarks or trade names used herein are the property of their respective owners. U.S. GOVERNMENT RIGHTS Use, duplication or disclosure by the US Government is subject to restrictions set forth in the OSIsoft, LLC license agreement and/or as provided in DFARS 227.7202, DFARS 252.227-7013, FAR 12-212, FAR 52.22719, or their successors, as applicable. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, mechanical, photocopying, recording or otherwise, without the written permission of OSIsoft, LLC. Published: April 17, 2020

2 | P a g e

Table of Contents

Contents

Table of Contents .............................................................................................................................................. 3 1. Introduction................................................................................................................................................. 5

1.1 Overview of Lab .................................................................................................................................. 5 1.2 Key Tasks ........................................................................................................................................... 5 1.3 Key Resources for Improving Analysis Service Performance .............................................................. 6 2. Directed Activity - Export Analytics Performance Data and Analyze ........................................................... 7 2.1 Objective of Activity ............................................................................................................................. 7 2.2 Activity Tasks ...................................................................................................................................... 7 2.3 Step by Step Explanation .................................................................................................................... 7 2.4 Solution ............................................................................................................................................. 15 2.5 Summary ........................................................................................................................................... 16 3. Directed Activity ? Build Error Checking into an Analytic........................................................................... 18 3.1 Objective of Activity ........................................................................................................................... 18 3.2 Tasks ................................................................................................................................................ 18 3.3 Step by Step Explanation .................................................................................................................. 18 3.4 Summary ........................................................................................................................................... 22 4. Directed Activity ? Pre-calculate Variables used in Multiple Analytics ....................................................... 23 4.1 Objective of Activity ........................................................................................................................... 23 4.2 Identify the Tasks .............................................................................................................................. 23 4.3 Step by Step Explanation .................................................................................................................. 23 5. Directed Activity ? Shift Calculations to PI Server ..................................................................................... 26 5.1 Objective of Activity ........................................................................................................................... 26 5.2 Tasks ................................................................................................................................................ 26 5.3 Step by Step Explanation .................................................................................................................. 27 6. Directed Activity ? Use Exit() Function to Optimize Conditional Calculations ............................................ 28 6.1 Objective of Activity ........................................................................................................................... 28 6.2 Tasks ................................................................................................................................................ 29 6.3 Step by Step Explanation .................................................................................................................. 29 7. Review ..................................................................................................................................................... 31 7.1 Analysis of Performance Statistics Following Fixes ........................................................................... 31 8. Optional Extra ? Manually Recalculate Analytics with Out of Order Data as Inputs................................... 33

8.1 Objective of Activity ...........................................................................................................................33 8.2 Tasks.................................................................................................................................................33 8.3 Steps .................................................................................................................................................33 Save the Date! ....................................................................................................................... Error! Bookmark not defined.

4 | P a g e

1. Introduction

1.1 Overview of Lab

A modern PI System uses an asset based approach to data analysis as opposed to the older tag based approach previously used. With PI Vision, the PI Integrators, and other tools that are asset focused there is great value to be had through an asset based PI System.

One of the biggest benefits comes from the Asset Analytics system that is part of PI AF. The "one-to-many" feature of creating analytics from templates and the greater power of the analytics engine have proven very powerful compared to the older Performance Equation system that was built into the PI Data Archive. The Asset Analytics engine is also much more scalable than the Performance Equation subsystem was, allowing for large numbers of Analytics.

There is a potential danger to PI System performance when analytics are created at large scales, however. Several factors contribute to potential performance issues, including:

? Badly written analytics that do not follow best practices

? Errors in source data causing slow performance in otherwise well-written analytics

? Analytics that are written in a way that works fine at small scales that are then scaled up to the point where they can cause issues

If there are too many performance issues with Asset Analytics, analytic results can be delayed or even skipped. This can lead to a lack of confidence among users in the results of the analytics.

1.2 Key Tasks

In this lab, we will cover how to improve the efficiency of a large-scale analytics implementation. We will start with how to analyze where to focus efforts in improving analytics. We will then show some common problems that could be causing performance issues with analytics along with potential improvements.

? Export Analysis Statistics from PSE ? Analyze the Analysis Statistics to identify the Analysis(es) that is consuming resources ? Add error checking to avoid Analyses in error ? Identify the best place and time to perform a calculation

? Scheduling ? Dependencies ? Calculate in AF or on the PI Data Archive ? Use the Exit() function to calculate only when necessary

? Monitor recalculations and recalculate manually if necessary

1.3 Key Resources for Improving Analysis Service Performance

1. KB Articles a. How to Troubleshoot PI Analysis Service Performance b. Backfilling and Recalculating Analyses with Asset Analytics c. Automatic Recalculation with the Analysis Service: Technical Details

2. Sebastien's blog on PI Square has a series on improving Analysis performance a. b. Use variables c. Data density and pattern d. Input attributes e. Analyses in warning or error f. Scheduling

6 | P a g e

2. Directed Activity - Export Analytics Performance Data and Analyze

2.1 Objective of Activity

When first building out a PI System with Analytics, it is rare to see performance problems due to the small number of analytics. When performance problems do occur, they are easy to isolate and improve. As the PI System is used and grows over time, and additional analytics are added and grow the system to a large scale, performance problems both become more likely and become harder to isolate. The first step to improving the performance of a PI System's Analytics is to identify where to focus troubleshooting efforts. In this activity we will show one method for identifying Analytics that could use attention.

2.2 Activity Tasks

? Export the Analytics performance data from PI System Explorer ? Use PowerShell to parse the data from XML format into a CSV format for import into Excel ? Import the data into Excel ? Enhance the data with additional derivative columns ? Analyze the data in a Pivot table

2.3 Step by Step Explanation

A. Export the performance data from PI System Explorer 1. Launch PI System Explorer 2. Go to the "Management" tab 3. Right click in the lower right under "Pending Operations"

4. Select "View Analysis Services Statistics" 8 | P a g e

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download