Administration Information PUBLIC

SAP Predictive Maintenance and Service, on-premise edition 1.0 FP03Document Version: 1.00 – 2017-08-18

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Content

1 Document History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2 Scheduling Regular Scoring of a Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

3 Loading Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83.1 Creating the Schema SAP_PDMS_REPLICATION_DATA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113.2 Creating a Procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

4 Deleting Data from the Data Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

5 Read Information About All Algorithms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

6 Creating Rules for Derived Signals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .176.1 Install and Configure the SAP HANA Rules Framework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176.2 Create Vocabulary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196.3 Create Rule Service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216.4 Create Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

7 Recommended Partitioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

8 Data Lifecycle Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

9 Migration of Artifacts Between Different Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279.1 Migrating Models Using SQL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28

10 Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

2 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Content

1 Document History

CautionBefore you read this document, make sure you have the latest version of this document. You can find the latest version at the following location:https://help.sap.com/viewer/d142f68ee0ef43298f4d00f2d73ea956/1.0.3/en-US

TipYou might need to refresh your browser to see the latest version of this document.

The following table provides an overview of the most important document changes:

Table 1:

Version Date Description

1.03 2017-08-18 Updated chapter:

Recommended Partitioning [page 24]

1.02 2017-07-03 Updated chapters:

● Scheduling Regular Scoring of a Model [page 4]

1.01 2017-03-24 Updated chapters:

● Loading Data [page 8]● Creating Rules for Derived Signals

[page 17]

1.00 2017-03-15 Initial Version

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Document History P U B L I C 3

2 Scheduling Regular Scoring of a Model

Prerequisites

● You have downloaded the latest SP of XS SERVICES 1 which contains the XSA Job Scheduler service, and installed the Job Scheduler service.For more information about how to install the XSA Job Scheduler service, see the SAP HANA Server Installation and Update Guide.

● The Space Developer role is assigned to the user who does the following tasks:○ Creating the service broker for the Job Scheduler service (Step 1)○ Creating the Job Scheduler service instance (Step 2)○ Binding an application to the Job Scheduler service (Step 3)

For more information about the Space Developer role and scheduling jobs, see the chapter Scheduling Jobs in XS Advanced.

Context

For more information about scheduling jobs in XS advanced, see the chapter Scheduling Jobs in XS Advanced in the SAP HANA Administration Guide.

Procedure

1. (If you have not registered a service broker)Configure the service broker for the Job Scheduler service as described in the chapter Configure the Service Broker for Job Scheduler.

NoteYou need to register a service broker once.

2. Create a Job Scheduler service instance in the XSA space you use for SAP Predictive Maintenance and Service, on-premise edition as described in the chapter Create a Job Scheduler Service Instance.

3. Bind the Job Scheduler service instance to an app of your choice as described in the chapter Bind an Application to the Job Scheduler Service.

This step is required to access the environment variables of the app datascience-service.

Fill in the variables as follows:

4 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Scheduling Regular Scoring of a Model

Table 2: Variables for Binding an App

Variable Value

<application name> datascience-service

<service instance name> Name of the Job Scheduler service instance that you cre­ated in Step 2

4. Extract the user name and password from the environment of the app datascience-service that you have bound the Job Scheduler service instance by executing the command xs env datascience-service.

After executing this command, the details of the Job Scheduler service instance are displayed as well.

Sample Code

"jobscheduler" : [ { "name" : "<Name of Job Scheduler service instance>", "label" : "jobscheduler", "tags" : [ "jobscheduler" ], "plan" : "default", "credentials" : { "password" : "<App environment password>", "user" : "<App environment user name>", "url" : "https://<Host of the Job Scheduler service>:<Port of the Job Scheduler service>" } }

The URL is used to access the Job Scheduler service, the user name and password are used to create a job using the Job Scheduler service.

5. Maintain jobs and job schedules as described in the chapter Maintain Jobs and Job Schedules in XS Advanced.

Sample Code

"name":"<Name of the job>", "description": "<Description of the job>", "action":"https://<Host of pdms router>:<Port of pdms router>/datascience/api/v1/models/score/<model ID>", "active": false, "httpMethod":"POST", "schedules": [ { "cron":"* * * * * * */10", "description": "This schedule runs every 10 seconds", "active": true, "startTime": {"date": "2016-01-01 00:00 +0000", "format": "YYYY-MM-DD HH:mm Z"} } ]}

NoteSo far, the example code creates only an inactive job without any authentication.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Scheduling Regular Scoring of a Model P U B L I C 5

In order to add basic authentication a second request (PUT /scheduler/jobs/:jobId) is necessary, as described in the chapter Job Scheduler REST API for XS Advanced.

Sample CodeSample code to configure a job

{ "active":true, "user":"<PDMS_DS_ADMIN>", "password":"<PASSWORD FOR PDMS_DS_ADMIN>", }

NoteThe basic authorization is now configured for the job which is also activated within the sample call. The handling of the X-CSRF token is taken care of by the Job Scheduler service, which is calling first GET on the the given URL to obtain the token, and uses the token to later do the original POST call. You can verify in the view SYS.M_TASKS that a new job task is created every 10 seconds.

For more information about the <PDMS_DS_ADMIN> user, see the guide Installation of SAP Predictive Maintenance and Service, on-premise edition.

NoteFor more information about how to work with the Job Scheduler service using REST APIs, see the chapter Job Scheduler REST API for XS Advanced.

6. Check and manage your jobs and job schedules in the Job Scheduler dashboard as described in the chapter The Job Scheduler Dashboard.

7. (If required) Stop a job schedule by updating the job configuration.a. Retrieve the job ID.

REST API call:

Sample Code

GET https://<Host of the Job Scheduler service>:<Port of the Job Scheduler service>/scheduler/jobs

This returns a list of jobs configured in the Job Scheduler service.b. Stop the job schedule.

REST API call:

Sample Code

PUT https://<Host of the Job Scheduler service>:<Port of the Job Scheduler service>/scheduler/jobs/<job _id>

with the following body

{ "active": false

6 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Scheduling Regular Scoring of a Model

}

This deactivates the job schedule.

Alternatively, you can use the Manage screen of the Job Scheduler dashboard to deactivate running jobs. For more information, see the chapter The Job Scheduler Dashboard.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Scheduling Regular Scoring of a Model P U B L I C 7

3 Loading Data

Prerequisites

● The role collection <pdms_executor_role> is assigned to the <PDMS_EXECUTOR_TECH_USER> user.This user is a technical user. The credentials of this user are required by the XSA Job Scheduler service to call back to the Executor Service app.For more information about role collections, see the chapters Maintaining Roles and Users in SAP HANA and Role Templates for SAP Predictive Maintenance and Service, on-premise edition in the guide Installation of SAP Predictive Maintenance and Service, on-premise edition 1.0.

● The following user-provided services are created in the space where the Executor Service application is deployed during the installation of SAP Predictive Maintenance and Service, on-premise edition 1.0○ executor-service-ups○ data-access-ups

For more information, see the Installation Guide for SAP Predictive Maintenance and Service, on-premise edition 1.0.

● The roles com.sap.pdms.sdm::DATA.Provider# and com.sap.pdms.sdm::DATA.Consumer# are assigned to the <User to insert data into the READINGS table> that loads data using the Executor Service app.For more information about roles, see the Security Guide for SAP Predictive Maintenance and Service, on-premise edition 1.0.

● The data model is set up.For more information about the data model, see the guide Installation of SAP Predictive Maintenance and Service, on-premise edition 1.0 → Installation of SAP Predictive Maintenance and Service, on-premise edition → Data Model.

● The schema SAP_PDMS_REPLICATION_DATA is owned by the <PDMS_TECH_USER> and adapted as explained in the chapter Creating the Schema SAP_PDMS_REPLICATION_DATA [page 11].

● Procedures are created by the <PDMS_TECH_USER> in the schema SAP_PDMS_REPLICATION_DATA as described in the chapter Creating a Procedure [page 12].

Context

Before users actually start using SAP Predictive Maintenance and Service, on-premise edition 1.0, data needs to be loaded to the following views:

● READINGS_T● LOCATIONS_T● MASTER_DATA

8 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Loading Data

Procedure

1. (If you have not registered a service broker) Configure the service broker for the Job Scheduler service as described in the chapter Configure the Service Broker for Job Scheduler.

NoteYou need to register a service broker once.

2. Create the Job Scheduler service instance pdms-jobscheduler in the XSA space you use for the Executor Service application as described in the chapter Create a Job Scheduler Service Instance.

3. (If you have installed the Executor Service application before you have installed the Job Scheduler service) Bind the Job Scheduler service instance to the Executor Service app as described in the chapter Bind an Application to the Job Scheduler Service.

Fill in the variables as follows:

Table 3: Variables for Binding an App

Variable Value

<application name> executor-service-backend

<service instance name> pdms-jobscheduler

4. Open the URL https://<hostname>:<port>/app/launchpad/index.html.

5. Log on to the launchpad with <PDMS_TECH_USER>.

6. Open the Executor Service app.7. Fill in the required entries.

Table 4: Entry Fields

Field Description

Name Unique name of the replication job

Description Description of the replication job

Procedure Name Name of the data replication procedure, which is located in the SAP_PDMS_REPLICATION_DATA schema.

For more information, see the chapters Creating the Schema SAP_PDMS_REPLICATION_DATA [page 11] and Creating a Procedure [page 12].

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Loading Data P U B L I C 9

Field Description

Schedule Schedule type

You can choose among the following schedule types:○ ONE_TIME

A non-recurring job that is executed only once at a spe­cific time

○ CRONA recurring job that is executed automatically at a spe­cific day and time

○ REPEAT_ATA recurring job that is repeated at an exact time

○ INTERVALA recurring job that is continuously repeated after a given interval

timeOrCron Recurrence pattern of the job

This field can have different formats depending on the schedule type you chose.

Example○ Human readable time such as Now, In 2

minutes, or Tomorrow at 4:00 p.m.○ Cron format such as 2015 * * fri 12 0 0

(Runs the job every Friday in 2015 at 12:00)For more information about Cron, see the Glossary [page 31].

○ Date format such as 2013-02-08 09+07:00

Description of the job schedule

8. To pass input parameters to your replication procedure, add Parameters.

The parameters are key-value pairs. The value is passed as an input parameter to the procedure when the procedure is called. The parameters are passed in the order in which they are added.

Example

Table 5: Parameters

Key Value

metricId LINE_CURRENT

retentionPeriod 123456789

9. Save the job.

10 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Loading Data

When your job is triggered, you can see the result in the XS Job Dashboard.

NoteTo access the XSA Job Dashboard, the required roles for running the Job Scheduler are assigned to your user.

In addition, the SpaceDeveloper role needs to be assigned to your user:

Sample Code

xs set-space-role <Your user> <Org> <Space> SpaceDeveloper

For more information about the XSA Job Dashboard, see the chapter The Job Scheduler Dashboard in the SAP HANA Administration Guide.

3.1 Creating the Schema SAP_PDMS_REPLICATION_DATA

Prerequisites

The schema is created and owned by the <PDMS_TECH_USER>. For more information, see the chapter Installing SAP Predictive Maintenance and Service, on-premise edition 1.0 in the Installation of SAP Predictive Maintenance and Service, on-premise edition 1.0.

Context

To load data for usage in SAP Predictive Maintenance and Service, on-premise edition 1.0, you need to adapt the schema SAP_PDMS_REPLICATION_DATA.

Procedure

1. Open the SAP HANA SQL console.2. Add the following permissions to the _Sys_Repo user:

GRANT SELECT ON <virtual source table from which data is loaded> TO _Sys_Repo; GRANT INSERT, UPDATE, DELETE ON <target table to which data is loaded> TO _Sys_Repo; GRANT SELECT ON SCHEMA SAP_PDMS_REPLICATION_DATA TO _Sys_Repo WITH GRANT OPTION;

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Loading Data P U B L I C 11

For more information about how to work with virtual tables, see the chapter Creating Virtual Tables in the SAP HANA Administration Guide.

Related Information

Loading Data [page 8]

3.2 Creating a Procedure

Prerequisites

You have created the schema SAP_PDMS_REPLICATION_DATA as explained in the chapter Creating the Schema SAP_PDMS_REPLICATION_DATA [page 11].

Context

Before you can load data for usage in SAP Predictive Maintenance and Service, on-premise edition 1.0, you need to create a data replication procedure.

You can create a procedure by executing SQL statements in the SAP HANA SQLScript editor.

For more information about SQL procedures, see the chapter Developing Procedures in the SAP HANA Developer Guide for SAP HANA Studio.

Procedure

1. Log on to SAP HANA studio with the<User to create replication procedures for the executor-service>.

2. Create a repository workspace as explained in Maintain a Repository Workspace in the SAP HANA Developer Guide for SAP HANA Studio.

3. In the SAP HANA Web-based Development Workbench Editor tool, create a repository package as explained in Create Repository Packages in the SAP HANA Developer Guide for SAP HANA Web-Based Development Workbench.

4. Check out the package you just created.5. In SAP HANA studio, create and share a SAP HANA project as explained in Using SAP HANA Projects in the

SAP HANA Developer Guide for SAP HANA Studio.

12 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Loading Data

6. Assign to your user the following user role permission to enable semantic code completion:

○ sap.hana.xs.dt.base::restapi○ sap.hana.xs.ide.roles::Developer

7. Create your procedure as explain in Create and Edit Procedures in the SAP HANA Developer Guide for SAP HANA Studio.

For more information about the CREATE PRODCEDURE statement, see the chapter CREATE PROCEDURE Statement (Procedural) in the SAP HANA SQL and System Views Reference.

Related Information

Loading Data [page 8]

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Loading Data P U B L I C 13

4 Deleting Data from the Data Model

Prerequisites

The role com.sap.pdms.sdm::DATA.Provider is assigned to your user with which you log on to SAP HANA studio.

Context

Proceed as described in the following steps to delete data from the data model that is described in the installation guide of SAP Predictive Maintenance and Service, on-premise edition.

Procedure

1. Log on to SAP HANA studio.2. Delete data from tables and views using the DELETE statement as explained in the chapter DELETE

Statement (Data Manipulation) in the SAP HANA SQL and System Views Reference.a. The following SQL statement is an example of how to delete data from the READINGS_T table.

Sample Code

DELETE FROM READINGS_T WHERE Thing = '<ThingId>';

After executing the above statement, the rows containing the specified <ThingId> are deleted from the READINGS_T table.

a. The following SQL statement is an example of how to delete data from the WORKACTIVITY table.

Sample Code

DELETE FROM "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.WORKACTIVITY_T" WHERE "AssignedTo" = '<User_ID>' OR "ReportedBy" = '<User_ID>';

After executing the above statement, the rows containing the specified <User_ID> are deleted from the WORKACTIVITY table.

14 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Deleting Data from the Data Model

5 Read Information About All Algorithms

Get information about algorithms.

Request

Permission: The following role collection is assigned to the <PDMS_DS_ADMIN> user:

● <pdms_datascience_role>

For more information about role collections, see the chapters Maintaining Roles and Users in SAP HANA and Role Templates for SAP Predictive Maintenance and Service, on-premise edition in the guide Installation of SAP Predictive Maintenance and Service, on-premise edition 1.0.

Format: JSON

URI: https://<hostname>:<router port>/datascience/api/v1/info

HTTP Method: GET

Response

Response Example

{ "rServerPlatform": "x86_64-pc-linux-gnu", "rVersion": "R version 3.3.1 (2016-06-21)", "rNickname": "Bug in Your Hair", "pdmsRVersion": "1.3.0", "packageStatuses": [ { "namespace": "com.sap.pdms.datascience", "dataScienceService": "PCA", "rPackageVersion": "com.sap.pdms.datascience.pca 1.3.0", "rTrainFunction": "FOUND PCAtrain", "rScoreFunction": "FOUND PCAscore" }, { "namespace": "com.sap.pdms.datascience", "dataScienceService": "RUL", "rPackageVersion": "com.sap.pdms.datascience.rul.weibull 1.3.0", "rTrainFunction": "FOUND Weibulltrain", "rScoreFunction": "FOUND Weibullscore" }, { "namespace": "com.sap.pdms.datascience", "dataScienceService": "EMD", "rPackageVersion": "com.sap.pdms.datascience.dbfa.emd 1.3.0", "rTrainFunction": "FOUND EMDtrain", "rScoreFunction": "FOUND EMDscore" }, {

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Read Information About All Algorithms P U B L I C 15

"namespace": "com.sap.pdms.datascience", "dataScienceService": "MAR", "rPackageVersion": "com.sap.pdms.datascience.mar 1.3.0", "rTrainFunction": "FOUND MARtrain", "rScoreFunction": "FOUND MARscore" }, { "namespace": "com.sap.pdms.datascience", "dataScienceService": "TEC", "rPackageVersion": "com.sap.pdms.datascience.tec 1.3.0", "rTrainFunction": "FOUND TECtrain", "rScoreFunction": "FOUND TECscore" } ]}

16 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Read Information About All Algorithms

6 Creating Rules for Derived Signals

Context

Proceed as described in the following steps to create rules for derived signals. The derived signals that are then generated based on the rules can be displayed in the insight provider for derived signals in the Asset Health Control Center and on the Asset Health Fact Sheet.

Procedure

1. Install and Configure the SAP HANA Rules Framework [page 17]2. Create Vocabulary [page 19]3. Create Rule Service [page 21]4. Create Rules [page 22]

6.1 Install and Configure the SAP HANA Rules Framework

Prerequisites

The schema SAP_PDMS_REPLICATION_DATA exists.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Creating Rules for Derived Signals P U B L I C 17

Context

To install and configure the SAP HANA Rules Framework (HRF), proceed as described in the following steps:

Procedure

1. Add the administrator role of the SAP HANA lifecycle manager to the SYSTEM user:

a. In the navigation tree of the new system, go to Security Users SYSTEM .b. Open the context menu of the SYSTEM user and then choose the option Open.c. In the Granted Roles tab, click (Add).d. Search for the role sap.hana.xs.lm.roles::Administrator and click OK.

e. Click File Save .2. Import the HRF delivery unit file HCORULEFW.tgz using the SAP HANA lifecycle manager:

a. In your browser, open the SAP HANA lifecycle manager: http://<full_domain_name>:80<instance_number>/sap/hana/xs/lm.

b. Log on using the same user and password you defined earlier in the Connection Properties window when you created the SAP HANA system.

c. Under Home, click Delivery Units.

d. Click Import .e. In the Import Delivery Unit dialog box, browse to and select the HCORULEFW.tgz file.f. Import the HCORULEFW.tgz file.g. Check the Notifications bar to ensure that the import is done successfully.

3. Add the following roles to the SYSTEM user in SAP HANA studio following the procedure described in Step 1.

○ sap.hrf.role.model::HrfAdmin○ sap.hrf.role.model::HrfRuleConsumer○ sap.hrf.role.model::HrfRuleViewer○ sap.hana.uis.db::SITE_DESIGNER○ sap.hana.uis.db::SITE_USER

4. Using any REST client, run the configuration script using the following request:a. URL: http://<full_domain_name>:80<instance_number>/sap/hrf/service/configurationb. Select the POST method.

Request body parameters:

Source Code

{ "timeZoneCode": "<ID>", "webApplicationConfigurations": "default"}Example: {"timeZoneCode": "005","webApplicationConfigurations": "default"}

18 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Creating Rules for Derived Signals

5. Grant permission on the schema created for derived signals:

Sample Code

grant SELECT, EXECUTE, INSERT, UPDATE on schema "SAP_PDMS_REPLICATION_DATA" to _SYS_REPO with grant option;

Related Information

Creating Rules for Derived Signals [page 17]Create Vocabulary [page 19]

6.2 Create Vocabulary

Context

To create vocabulary and a vocabulary schema, proceed as described in the following steps:

Procedure

1. (Optional) Deploy the SAP HANA Rules Framework modeling tools in SAP HANA studio.

This is a one-time task to install the add-on for vocabulary creation in SAP HANA studio.

a. Open SAP HANA studio.

b. Choose Help Install New Software .c. In the Install window, click Add.d. In the Add Repository window, enter a name and location for your repository.

For the repository location, enter http://<hostname>:80<instance_number>/sap/hrf/updatesite.

e. Enter the logon credentials of the SAP HANA machine.f. Select SAP HANA Rules Framework - Modeling Tools.g. Click Next and Next again.h. After you have read and accepted the terms of the license agreement, click Finish.i. Click OK when you are asked install software with unsigned content.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Creating Rules for Derived Signals P U B L I C 19

j. When the installation has completed, restart SAP HANA studio.2. Create the XS project SAP_PDMS_REPLICATION_DATA in the SAP HANA studio Project Explorer view.

3. Select the project and follow the path File New Other .

4. Select SAP HANA Operational Process Intelligence Vocabulary .5. Click Next and enter DS_VOCAB as name, and click Finish.

6. Create the table DERIVED_SIGNALS_METRIC_DATA in the SAP_PDMS_REPLICATION_DATA schema:

Source Code

create column table "SAP_PDMS_REPLICATION_DATA"."DERIVED_SIGNALS_METRIC_DATA"("ThingType" NVARCHAR(81), "PropertySetTypeID" NVARCHAR(30),"PropertyID" NVARCHAR(30),"Timestamp" TIMESTAMP,"Value" DOUBLE,"AssetThingID" NVARCHAR(32),"ComponentThingID" NVARCHAR(32),"AssetName" NVARCHAR(255),"ComponentName" NVARCHAR(255));

7. Add the following fields to the data object DERIVED_SIGNALS_METRIC_DATA:

Table 6: Data Object Fields

Field Type

ThingType NVARCHAR(81)

PropertySetTypeID NVARCHAR(30)

PropertyID NVARCHAR(30)

Timestamp TIMESTAMP

Value DOUBLE

AssetThingID NVARCHAR(32)

ComponentThingID NVARCHAR(32)

AssetName NVARCHAR(255)

ComponentName NVARCHAR(255)

8. Add the output type DERIVED_SIGNALS_METRIC_DATA_OUTPUT..

9. Add the following fields to the output type, and activate the vocabulary:

Table 7: Parameters

Field Type Length

SEVERITY NVARCHAR 50

ALERT_TYPE NVARCHAR 50

20 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Creating Rules for Derived Signals

Table 8: Static Parameters

Field Expression

VALUE DERIVED_SIGNALS_METRIC_DATA.VALUE

Related Information

Creating Rules for Derived Signals [page 17]Create Rule Service [page 21]

6.3 Create Rule Service

Context

To create the rule service needed to create rules for derived signals, proceed as described in the following steps:

Procedure

1. Access the SAP HANA Rules Framework web application for rule services and rules in your web browser: http://<full_domain_name>:80<instance_number>/sap/hana/uis/clients/ushell-app/shells/fiori/FioriLaunchpad.html?siteId=sap.hrf.ui|app|app.

2. Click on the Rule Service tile and create a new rule service with the following values:

Table 9: Entries to Create Rule Service

Field Value

Name PDMS_DS_DATA

Description PDMS_DS_DATA

Consumption Artifact Procedure and View

Vocabulary SAP_PDMS_REPLICATION_DATA::DS_VOCAB

Resulting Data Object DERIVED_SIGNALS_METRIC_DATA

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Creating Rules for Derived Signals P U B L I C 21

Field Value

Keys ThingType, PropertySetTypeID, PropertyID, Timestamp, ComponentThingID

Output DERIVED_SIGNALS_METRIC_DATA_OUTPUT

Rule Assignment Automatic

Related Information

Creating Rules for Derived Signals [page 17]Create Rules [page 22]

6.4 Create Rules

Context

To create rules for derived signals, proceed as described in the following steps:

Procedure

1. Access the SAP HANA Rules Framework web application for rule services and rules in your web browser: http://<full_domain_name>:80<instance_number>/sap/hana/uis/clients/ushell-app/shells/fiori/FioriLaunchpad.html?siteId=sap.hrf.ui|app|app.

2. Click on the Rules tile and create a new rule with the following values:

Table 10: Entries to Create Rule

Field Value

Name DSRULE

Description DSRULE

Vocabulary SAP_PDMS_REPLICATION_DATA::DS_VOCAB

22 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Creating Rules for Derived Signals

Field Value

Condition ThingType of the DERIVED_SIGNALS_METRIC_DATA is equal to 'com.sap.pdms.example.trains:Battery' and PropertySetTypeID of the DERIVED_SIGNALS_METRIC_DATA is equal to 'power' and PropertyID of the DERIVED_SIGNALS_METRIC_DATA is equal to 'Current' and Value of the DERIVED_SIGNALS_METRIC_DATA is greater than 0.6 and ComponentThingID of the DERIVED_SIGNALS_METRIC_DATA is equal to 'LOCOMOTIVE-SP02-2500-BT'

3. Click on Add Output, select DERIVED_SIGNALS_METRIC_DATA_OUTPUT, and add values such as the following:

Table 11: Derived Signals Values

Field Value

SEVERITY 'CRITICAL'

ALERT_TYPE 'SAFETY'

Related Information

Creating Rules for Derived Signals [page 17]

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Creating Rules for Derived Signals P U B L I C 23

7 Recommended Partitioning

As part of performance, multi-level partitioning is recommended for the following tables:

● "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.READINGS_T.0"● "SAP_PDMS_DATA". "com.sap.pdms.sdm::DATA.SCORES_T"

At the first level, hash partitioning is used on Thing and Thing Types columns with X buckets. On the second level, range partitioning is used to timestamp column ("TimeStamp 0" for READINGS_T.0 and "TimeStamp" for SCORES_T) with Y interval buckets.

For example, consider the data available in 2016:

ALTER TABLE "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.READINGS_T.0" PARTITION BY HASH("Thing", "ThingType") PARTITIONS 5,RANGE(month("Timestamp.0")) (PARTITION '2016-01' <= values < '2016-04', PARTITION '2016-04' <= values < '2016-07', PARTITION '2016-07' <= values < '2016-10', PARTITION '2016-10' <= values < '2017-01', PARTITION OTHERS)

ALTER TABLE "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.SCORES_T" PARTITION BY HASH("Thing", "ThingType") PARTITIONS 5,RANGE(month("Timestamp")) (PARTITION '2016-01' <= values < '2016-04', PARTITION '2016-04' <= values < '2016-07', PARTITION '2016-07' <= values < '2016-10', PARTITION '2016-10' <= values < '2017-01', PARTITION OTHERS)

24 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Recommended Partitioning

8 Data Lifecycle Management

Deleting Data from Tables READINGS_T and SCORES_T

You need to decide on a retention period for created procedures, tables, and table types. You can schedule regular replication procedures using the Executor Service app on the launchpad and the SAP HANA XSA Job Scheduler service.

For more information about the app and the service, see the chapter Loading Data [page 8].

TipTo delete data from the READINGS_T table and SCORES_T table, we recommend performing the deletion as part of a replication procedure.

You should also run the following commands to optimize compression after replication:

alter table "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.READINGS_T.0" series reorganize

alter table "SAP_PDMS_DATA"."com.sap.pdms.sdm::DATA.SCORES_T.0" series reorganize

For more information about handling series data, see the SAP HANA Series Data Developer Guide.

TipTo support housekeeping of the machine learning engine, the procedure SAP_PDMS_DSS.CLEANUP_BUT_N with an integer parameter is provided. Calling this procedure with a certain number will keep the tables, procedures, and other database artifacts, as well as log entries for the specified number of jobs.

ExampleCalling SAP_PDMS_DSS.CLEANUP_BUT_N(10) will keep the latest 10 jobs.

To delete the trained binary model data for models that have been deleted, you can use the following statement:

DELETE FROM SAP_PDMS_DSS.TRAIN_OUT WHERE MODEL_ID IN ( SELECT MODEL_ID FROM SAP_PDMS_DSS.MODEL_MASTER WHERE IS_DELETED=TRUE);

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Data Lifecycle Management P U B L I C 25

Avoiding Running Out of Memory

● Only replicate data into SAP HANA that really has to be shown in SAP Predictive Maintenance and Service, on-premise edition.

● Delete data that is no longer needed. Keep the retention period for data in SAP HANA as short as possible.● If applicable, apply temporal aggregation during the replication of data into SAP HANA.

ExampleThe raw data may consist of measurements taken every second, but only hourly measurements are relevant to users of SAP Predictive Maintenance and Service, on-premise edition. In this case, raw measurements can be aggregated to one measurement per hour, thus saving space in SAP HANA.

26 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Data Lifecycle Management

9 Migration of Artifacts Between Different Systems

You might want to migrate your implementation of SAP Predictive Maintenance and Service, on-premise edition from your development environment to your text environment to your production environment. The way how artifacts are migrated, depends on their type of content.

Configuration Artifacts

To export and import configuration artifacts such as the Thing model configuration or the configuration of insight providers, you can use REST APIs.

SAP HANA Content

You can migrate SAP HANA content such as the following:

● Stored procedures used for key figures or for data replication using the Executor Service app● Views created to manage the machine learning engine (scoring and training of algorithms)● Rule artifacts generated by the SAP HANA Rules Framework and strored in the SAP HANA repository

SAP HANA content can be migrated by using the transport mechanism provided by SAP HANA such as delivery units. For more information about migrating SAP HANA content using the SAP HANA studio, see Export and Import feature in SAP HANA Studio .

Content of the Machine Learning Engine

Models of the machine learning engine are BLOBs in the SAP_PDMS_DSS.TRAIN_OUT table. Thus, models can be migrated like other database content. However, the model ID cannot be copied because this GUID is unique only per system. Copying the model ID and migrating it to another system might, thus, result in conflicts.

To migrate a model, the model configuration, model meta data on training, and the trained model are required. For a detailed description of how to migrate models, see the chapter Migrating Models Using SQL [page 28].

Data

You can migrate data such as things, thing hierarchies, time series, or master data by exporting and importing them using the SAP HANA mechanisms such as binary and CSV.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Migration of Artifacts Between Different Systems P U B L I C 27

For more information about CSV exports and imports, and binary exports and imports, watch the following videos:

SAP HANA Academy - Exporting to CSV

SAP HANA Academy - Importing Data from CSV file

SAP HANA Academy - Exporting to Binary

9.1 Migrating Models Using SQL

Context

To migrate models between system, follow the steps described below.

Procedure

1. In the source system, before starting the export, execute the following statements as SYSTEM user in SAP HANA studio:

CREATE LOCAL TEMPORARY COLUMN TABLE #PARAMS LIKE "_SYS_DI"."TT_PARAMETERS"; CREATE LOCAL TEMPORARY COLUMN TABLE #PRIVS LIKE "_SYS_DI"."TT_SCHEMA_PRIVILEGES";INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'SELECT', 'PDMS_DS_ADMIN', NULL);CALL _SYS_DI.GRANT_CONTAINER_SCHEMA_PRIVILEGES('SAP_PDMS_DSS', #PRIVS, #PARAMS, ?, ?, ? );DROP TABLE #PARAMS;DROP TABLE #PRIVS;

2. As <PDMS_DS_ADMIN> user, execute the following statements to set the working schema:

NotePlease replace <--condition goes here> by a filter condition to filter for the models you want to migrate.

SET SCHEMA "SAP_PDMS_DSS_FUSION_SCHEMA"; DO BEGIN models = SELECT MODEL_ID FROM SAP_PDMS_DSS.MODEL_MASTER WHERE <--condition goes here>; CREATE COLUMN TABLE TRANSPORT_MODEL_MASTER AS ( SELECT * FROM SAP_PDMS_DSS.MODEL_MASTER WHERE MODEL_ID IN (SELECT * FROM :models)

28 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Migration of Artifacts Between Different Systems

); CREATE COLUMN TABLE TRANSPORT_MODEL_COLUMNS_CONFIG AS ( SELECT * FROM SAP_PDMS_DSS.MODEL_COLUMNS_CONFIG WHERE MODEL_ID IN (SELECT * FROM :models) ); CREATE COLUMN TABLE TRANSPORT_MODEL_PARAMETERS AS ( SELECT * FROM SAP_PDMS_DSS.MODEL_PARAMETERS WHERE MODEL_ID IN (SELECT * FROM :models) ); CREATE COLUMN TABLE TRANSPORT_MODEL_VERSIONS AS ( SELECT * FROM SAP_PDMS_DSS.MODEL_VERSIONS A WHERE A.MODEL_ID IN (SELECT * FROM :models) AND A.MODEL_VERSION = (SELECT MAX(MODEL_VERSION) FROM SAP_PDMS_DSS.MODEL_VERSIONS B WHERE B.MODEL_ID = A.MODEL_ID) ); CREATE COLUMN TABLE TRANSPORT_GLOBAL_CONFIG AS ( SELECT * FROM SAP_PDMS_DSS.GLOBAL_CONFIG WHERE (DSS_TYPE,NAMESPACE) IN ( SELECT DISTINCT DSS_TYPE,NAMESPACE FROM SAP_PDMS_DSS.MODEL_MASTER WHERE MODEL_ID IN (SELECT MODEL_ID FROM :models) ) ); CREATE COLUMN TABLE TRANSPORT_MODEL_BLOB AS ( SELECT * FROM SAP_PDMS_DSS.TRAIN_OUT A WHERE A.MODEL_ID IN (SELECT * FROM :models) AND A.MODEL_VERSION = (SELECT MAX(MODEL_VERSION) FROM SAP_PDMS_DSS.MODEL_VERSIONS B WHERE B.MODEL_ID = A.MODEL_ID) );END;

3. After export, reset the permissions as the SYSTEM user with the following statements:

CREATE LOCAL TEMPORARY COLUMN TABLE #PARAMS LIKE "_SYS_DI"."TT_PARAMETERS"; CREATE LOCAL TEMPORARY COLUMN TABLE #PRIVS LIKE "_SYS_DI"."TT_SCHEMA_PRIVILEGES";INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'SELECT', 'PDMS_DS_ADMIN', NULL);CALL _SYS_DI.REVOKE_CONTAINER_SCHEMA_PRIVILEGES('SAP_PDMS_DSS', #PRIVS, #PARAMS, ?, ?, ? );DROP TABLE #PARAMS;DROP TABLE #PRIVS;

4. Transport all tables "SAP_PDMS_DSS_FUSION_SCHEMA"."TRANSPORT%" to the target system using the SAP HANA transport mechanisms.

NoteMake sure the TRANSPORT% tables are written in schema SAP_PDMS_DSS_FUSION_SCHEMA on the target system. If you use a different schema name, please replace the schema SAP_PDMS_DSS_FUSION_SCHEMA with your schema name when executing the statements in the following steps.

5. Switch to the target system in SAP HANA studio.6. In the target system, before starting the import, execute the following statements as SYSTEM user to give the

PDMS_DS_ADMIN permission to insert and update tables in the schema SAP_PDMS_DSS:

CREATE LOCAL TEMPORARY COLUMN TABLE #PARAMS LIKE "_SYS_DI"."TT_PARAMETERS"; CREATE LOCAL TEMPORARY COLUMN TABLE #PRIVS LIKE "_SYS_DI"."TT_SCHEMA_PRIVILEGES";INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'INSERT', 'PDMS_DS_ADMIN', NULL);INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'UPDATE', 'PDMS_DS_ADMIN', NULL);

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Migration of Artifacts Between Different Systems P U B L I C 29

CALL _SYS_DI.GRANT_CONTAINER_SCHEMA_PRIVILEGES('SAP_PDMS_DSS', #PRIVS, #PARAMS, ?, ?, ? );DROP TABLE #PARAMS;DROP TABLE #PRIVS;

7. As <PDMS_DS_ADMIN>, execute the following statements to update tables in the schema SAP_PDMS_DSS_FUSION_SCHEMA:

SET SCHEMA "SAP_PDMS_DSS_FUSION_SCHEMA"; DO BEGIN INSERT INTO SAP_PDMS_DSS.MODEL_MASTER SELECT * FROM TRANSPORT_MODEL_MASTER; INSERT INTO SAP_PDMS_DSS.MODEL_COLUMNS_CONFIG SELECT * FROM TRANSPORT_MODEL_COLUMNS_CONFIG; INSERT INTO SAP_PDMS_DSS.MODEL_PARAMETERS SELECT * FROM TRANSPORT_MODEL_PARAMETERS; INSERT INTO SAP_PDMS_DSS.MODEL_VERSIONS SELECT * FROM TRANSPORT_MODEL_VERSIONS; UPSERT SAP_PDMS_DSS.GLOBAL_CONFIG SELECT * FROM TRANSPORT_GLOBAL_CONFIG; INSERT INTO SAP_PDMS_DSS.TRAIN_OUT SELECT * FROM TRANSPORT_MODEL_BLOB;END;

8. After import, as SYSTEM user, execute the following statements to reset the permissions for the <PDMS_DS_ADMIN> in the target system:

CREATE LOCAL TEMPORARY COLUMN TABLE #PARAMS LIKE "_SYS_DI"."TT_PARAMETERS"; CREATE LOCAL TEMPORARY COLUMN TABLE #PRIVS LIKE "_SYS_DI"."TT_SCHEMA_PRIVILEGES";INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'INSERT', 'PDMS_DS_ADMIN', NULL);INSERT INTO #PRIVS (PRIVILEGE_NAME, PRINCIPAL_NAME, PRINCIPAL_SCHEMA_NAME) VALUES ( 'UPDATE', 'PDMS_DS_ADMIN', NULL);CALL _SYS_DI.REVOKE_CONTAINER_SCHEMA_PRIVILEGES('SAP_PDMS_DSS', #PRIVS, #PARAMS, ?, ?, ? );DROP TABLE #PARAMS; DROP TABLE #PRIVS;

Related Information

Migration of Artifacts Between Different Systems [page 27]

30 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Migration of Artifacts Between Different Systems

10 Glossary

Table 12: Term Definitions

Term Definition

Cron Cron is a Unix program that can be used to execute recurrent jobs automatically on a specific day and at a specific time.

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Glossary P U B L I C 31

Important Disclaimers and Legal Information

Coding SamplesAny software coding and/or code lines / strings ("Code") included in this documentation are only examples and are not intended to be used in a productive system environment. The Code is only intended to better explain and visualize the syntax and phrasing rules of certain coding. SAP does not warrant the correctness and completeness of the Code given herein, and SAP shall not be liable for errors or damages caused by the usage of the Code, unless damages were caused by SAP intentionally or by SAP's gross negligence.

AccessibilityThe information contained in the SAP documentation represents SAP's current view of accessibility criteria as of the date of publication; it is in no way intended to be a binding guideline on how to ensure accessibility of software products. SAP in particular disclaims any liability in relation to this document. This disclaimer, however, does not apply in cases of willful misconduct or gross negligence of SAP. Furthermore, this document does not result in any direct or indirect contractual obligations of SAP.

Gender-Neutral LanguageAs far as possible, SAP documentation is gender neutral. Depending on the context, the reader is addressed directly with "you", or a gender-neutral noun (such as "sales person" or "working days") is used. If when referring to members of both sexes, however, the third-person singular cannot be avoided or a gender-neutral noun does not exist, SAP reserves the right to use the masculine form of the noun and pronoun. This is to ensure that the documentation remains comprehensible.

Internet HyperlinksThe SAP documentation may contain hyperlinks to the Internet. These hyperlinks are intended to serve as a hint about where to find related information. SAP does not warrant the availability and correctness of this related information or the ability of this information to serve a particular purpose. SAP shall not be liable for any damages caused by the use of related information unless damages have been caused by SAP's gross negligence or willful misconduct. All links are categorized for transparency (see: http://help.sap.com/disclaimer).

32 P U B L I COperation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0

Important Disclaimers and Legal Information

Operation Information for SAP Predictive Maintenance and Service, on-premise edition 1.0Important Disclaimers and Legal Information P U B L I C 33

go.sap.com/registration/contact.html

© 2017 SAP SE or an SAP affiliate company. All rights reserved.No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP SE or an SAP affiliate company. The information contained herein may be changed without prior notice.Some software products marketed by SAP SE and its distributors contain proprietary software components of other software vendors. National product specifications may vary.These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty.SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names mentioned are the trademarks of their respective companies.Please see http://www.sap.com/corporate-en/legal/copyright/index.epx for additional trademark information and notices.