IoT Tips

In this video we show the setup for anomaly detection (ThingWatcher) in release 8.4. We also show how to create an anomaly alert.  

How to score new data with ThingWorx Analytics ?   The following is valid starting with ThingWorx Analytics (TWA) 8.3.0   Overview   Once a training model has been created, one of the main objective is to score new data to predict the value for the goal ThingWorx Analytics can score new data in 2 ways: Batch scoring Real time scoring Batch scoring   Batch scoring will be used when a large amount of data needs to be scored. To perform a batch scoring we will usually follow steps similar to the below ones: Upload the historic data Create a new model with this historic data Upload new data – the one to be scored Perform a prediction job to score those new data Retrieve the prediction job result Uploading the new data can be done in different ways. If using a large amount of data, it can be easier to upload the data via a csv file in a similar way as the historic data. This is the way used in ThingWorx Analytics Builder. If the amount of data is more limited this can be sent in the body of the scoring request. The post Analytics: Prediction Methods Mashup  shows a good example of how to do this using the PredictionThing.BatchScore service. We are focusing below on ThingWorx Analytics Builder, that is uploading new data via a csv file. In order to perform the scoring job only on the new data in step 4 above, we need to be able to filter those added data. If the dataset has already suitable column/feature such as a timestamp for example, we can use this to score only new data after timestamp > newdate, assuming all data are in chronological order. If the dataset has no such feature, we will have to add one  beforehand when we first upload the historic data in step 1 above. We often use a new column/feature named record_purpose to this effect. So initial data can take a value of training for this record_purpose feature since they are used to create the initial model. Then new added data to be scored can get any value that identify those rows only. It is important to note that this record_purpose feature needs to be set with the optType INFORMATIONAL so as to not be taken into account by the learning algorithms.   The video below shows those steps while using ThingWorx Analytics Builder   Real time scoring   Real time scoring is better suited for small amount of data. The process for real time scoring can be done either via the Analytics Server PredictionThing RealTimeScore service or using the Analytics Manager framework. The posts How to work with ordinal and categorical data in ThingWorx Analytics  and Analytics: Prediction Methods Mashup do give  examples of the use of the RealTimeScore service.   We will concentrate below on the Analytics Manager. The process involves the following steps: In Analytics Manager Create an Analysis Provider that uses the AnalyticsServerConnector connector Publish the model created in ThingWorx Analytics Builder to Analytics Manager Enable the model created Create an Analysis Event Map the properties to the datashape field Enable the Event In ThingWorx Composer Relevant properties of the Thing used in the Analysis Event are updated in someway This trigger the analysis job to be executed The scoring result is populated into the result property mapped in the Analysis event The Help Center has got more detailed about this process. The following video shows those steps Following articles can also be of interest for this topic: How to use ThingPredictor in release 8.3 of ThingWorx Analytics Server ? Publish model from Analytics Builder into Analytics Manager using TW.AnalysisServices.AnalyticsServer.AnalyticsServerConnector Creating Template For Thing, And Configure Analysis Event For Real-Time Scoring via Analytics Manager Note that the AnalyticsServerConnector connector in release 8.3 replaces the ThingPredictor connector from previous releases.

Hi, ThingWorx users! We’re excited to share that we have partnered with InfluxData to make time series analysis in ThingWorx even easier. InfluxDB is a database by InfluxData that is “built specifically for metrics and events that empower developers to build next-generation IIoT, analytics and monitoring applications.”   Why InfluxData?  Today, application developers expect robust querying capabilities, fast response time, easy ways of aggregating and pivoting on data and leveraging results for reporting and visualization. IT and devops administrators also expect cost-effective storage and easy ways of aging data through archiving and the ability to keep large amounts of historical data to satisfy analysis requirements.   That’s why we’ve partnered with InfluxData to make it easier for developers to store, analyze and act on IIoT data in real-time. With InfluxData, developers can build connected IIoT applications more quickly while still incorporating the following capabilities: monitoring real-time alerting predictive maintenance streaming data anomaly and event detection visual and report-based analysis   We considered a few technologies for the purpose of improving ThingWorx time series analysis. Here are a few reasons we chose InfluxData: high compression of data ~45x ability to handle millions of writes per second* ability to read around thousands of rows in milliseconds* supports the standard time series functions of sampling, interpolation, time bucketing, aggregation, selector, transformation, predictor, etc.  * Query and write times will vary based on an individual ThingWorx application’s implementation with Influx. For example, as the number of concurrent reads increases, the query speed decreases. With the upcoming 8.4 release, the ThingWorx Sizing Guide will be updated to reflect representative performance for ThingWorx developers.   In addition to improved query capability, ThingWorx time series with Influx can now use less memory and CPU, giving your platform servers a bit of a break.   To start strategizing on how InfluxData can help you in your ThingWorx journey, here is a sneak preview of what it will look like:   New Features The new ThingWorx Influx Persistence Provider will make query services like ValueStream Thing QueryPropertyHistory, Stream Thing QueryStreamData and QueryStreamEntries even better. Simply create a new instance of the persistence provider, configure it to use your InfluxDB instance, create a new value stream (or stream) from the new persistence provider, and you’ll be writing, reading and analyzing your time series data like never before.   We’re also introducing a new enhancement to improve InfoTable support with time series data, including providing the ability to use a driver property. The driver property can be specified with the QueryPropertyHistoryWithDriverProperty service for time alignment and filling backward/forward in your stream queries.   Let’s walk through a driver property example where you have the properties of temperature, speed and battery level. Timestamp Temperature Speed Battery Level 1480589076592000000 80.003 5012 79 1480589077537000000 80.010 5011 79 1480589077550000000 80.010 5009 79 1480589077562000000 80.030 5011 78   Let’s say temperature is the key driver for your analysis. In other words, you are not concerned if speed or battery level changes—you only care about when temperature changes. We can specify temperature to be the driver property for that particular time and only return stream values for temperature, speed and battery level when temperature changes. If speed or battery level changes (but temperature does not change), the rows associated with those changes would not be included in the results set because neither speed nor battery level is a driver property. See chart below.  Timestamp Temperature (driver) Speed Battery Level 1480589076592000000 80.003 5012 79 1480589077537000000 80.010 5011 79 1480589077562000000 80.030 5011 78  Note that only three of the four rows are returned above because one entry in the original table did not have a change in temperature.    Stay Tuned Look out for these time series improvements and InfluxData integration in our upcoming 8.4 release. I’ll be sure to keep you updated on additional new features coming in our next release (like Orchestration and Mashup Builder 2.0), so check back shortly or subscribe to this Community so we can stay in touch. As always, if you have any questions, just ask Kaya!   Stay connected, Kaya

Based on Google's Spanner DB; CockroachDB is a distributed SQL DB scaling horizontally; surviving disk, machine, rack & even datacenter failures. It is built to automatically replicate, rebalance & recover with minimal configuration  See What is CockroachDB? for more.   Useful in use cases requiring: Distributed or replicated OLTP Multi-datacenter deployments Multi-region deployments Cloud migrations Cloud-native infrastructure initiatives Note: CockroachDB in current state isn't suitable for heavy analytics / OLAP.   Feature that makes it really attractive As mentioned above, scaling horizontally it requires minimal configuration out of the box allowing quick setup starting from local laptop/machine as shown below it can scale easily to single dedicated server, development/public cloud cluster. Due to easy setup, adding new nodes is as simple as starting the cockroach utility.See CockroachDB FAQ for more. To top it off, it uses PostgreSQL Wire protocol and PostgreSQL's dialect further reducing configuration and special JDBC driver requirements when a ThingWorx is configured with PostgreSQL as persistence provider.   Setting up cockroach DB cluster Download required binary or docker version from Install CockroachDB available for Mac, Linux & Windows   PS :Following setup uses Window's binary on a VM with Win10 64 bit, 6G RAM.     Starting Cluster node Open command prompt and navigate to the directory where cockroach.exe is unzipped, and launching the node with following command prompt     cockroach.exe start --insecure --host=10.128.13.183 --http-port=8082     This will start a node on defined host in insecure mode with its web based DB administration console on port 8082 and DB listening on default port 26257. Note it will log a security warning since node is started in insecure mode due to the tag --insecure, like so     * * WARNING: RUNNING IN INSECURE MODE! * * - Your cluster is open for any client that can access 10.128.13.183. * - Any user, even root, can log in without providing a password. * - Any user, connecting as root, can read or write any data in your cluster. * - There is no network encryption nor authentication, and thus no confidentiality. * * Check out how to secure your cluster: https://www.cockroachlabs.com/docs/stable/secure-a-cluster.html * CockroachDB node starting at 2018-03-16 11:52:57.164925 +0000 UTC (took 2.1s) build: CCL v1.1.6 @ 2018/03/12 18:04:35 (go1.8.3) admin: http://10.128.13.183:8082 sql: postgresql://root@10.128.13.183:26257?application_name=cockroach&sslmode=disable logs: C:\CockroachDb\cockroach116\cockroach-data\cockroach-data\logs store[0]: path=C:\CockroachDb\cockroach116\cockroach-data\cockroach-data status: restarted pre-existing node clusterID: 012d011e-acef-47e2-b280-3efc39f2c6e7 nodeID: 1     Ensure that the secure mode is used when deploying in production.   Starting 2 additional nodes   Starting node2 cockroach.exe start --insecure --store=node2 --host=10.128.13.183 --port=28258 --http-port=8083 --join=10.128.13.183:26257   Starting node 3   cockroach.exe start --insecure --store=node2 --host=10.128.13.183 --port=28259 --http-port=8084 --join=10.128.13.183:26257     Note: Both of these 2 nodes are joining the cluster via 10.128.13.183:26257 (port for the node 1)   Verifying the live cluster and nodes via the web based CockroachDB admin console Open a web browser with any of the above node's http-port e.g. http://10.128.13.183:8084 Click on the View nodes list on the right panel   This will open the nodes list page   Connecting to ThingWorx as external datastore Good news, if your ThingWorx is running with PostgreSQL persistence provider, then no additional JDBC driver needed as CockroachDB uses the PostgreSQL wire protocol additionally, the SQL dialect is that of PostgreSQL For any other persistence provider download and install the PostgreSQL Relational Database Connector from ThingWorx Marketplace.   Creating a database in the cluster Start SQL client connecting to any of the running node, open a command prompt navigate to the directory containing cockroach.exe use following command:   cockroach sql --insecure --port=26257 This will change the prompt to root@<serverName/IP>:26257> Since above command logs in insecure mode no password is needed, default admin username is root in CockroachDb, use following to create a database   create database thingworx; show databases; root@10.128.13.183:26257/> SHOW databases; +--------------------+ | Database | +--------------------+ | crdb_internal | | information_schema | | pg_catalog | | system | | thingworx | | thingworxdatastore | +--------------------+ (6 rows)   This confirms thingworx database is created Creating a user to access that database CREATE USER cockroach WITH PASSWORD 'admin'; This will grant all rights to "cockroach" user on the database thingworx database   grant all on database thingworx to cockroach; Creating a Thing & connecting to CockroachDB via ThingWorx Composer For below example ThingWorx is using PostgreSQL as persistence provider. Create a Thing based of Database Thing Template Use following connection settings:   JDBC Driver Class Name : org.postgresql.Driver JDBC Connection String : jdbc:postgresql://<serverIp/name>:26257/thingworx?sslmode=disable Database User Name : cockroach Database password : <password>   Navigate to the Properties to verify the connectivity   Creating table(s) Now that the Thing is connected to the database, there are following ways DB objects can be created Via Thing based SQL Command Via SQL CockroachDB's SQL client Following command will create a small demo table CREATE TABLE demo ( id INT, demovalue STRING) Use SQLCommand as JavaScript handler when using above statement to create table directly from ThingWorx's Database Thing Verifying the Database & a table created within that DB via the web admin console of CockroachDb Under the left panel click on the Databases from the home page of one of the node's web admin consloe e.g. http://localhost:8084     Apart from other useful information about the database e.g. the database size and total number of tables, etc., clicking on the table name will also show the sql used to create it (including the defaults).   Creating couple of Database Thing services to perform bulk insert into the table from ThingWorx Composer Insert Service created as SQL Command with code snippet, service takes 2 inputs of type int and string   insert into demo values ([[id]], [[demoValue]]) JavaScript service executing bulk demo data insert by wrapping the SQL service created above   for (i=0; i<2000; i++) { var params = { id: i /* INTEGER */, demoValue: 'Insert from node 0 while node 3 is down' /* STRING */ }; // result: NUMBER var result = me.InsertDemo(params); }   At this point different users in ThingWorx with sufficient access rights can create their DB Things in ThingWorx Composer and can use any of the node address to read/write the data to the CockroachDB cluster. For the purpose of demo one node was stopped while other 2 were running and data was written to the clsuter via the test service created above. Once the 3rd node was restarted we can see the automatic replication happening between the nodes; this can be seen live via the web based admin console of any of the running node's web console.   As highlighted above at certain point in time after i.e 1500hrs all nodes were synced with the data, including the node3 ,which as mentioned above was down while data was being inserted to the cluster. All of the above replication process was done using default configuration.  

  Remotely administer Windows Edge IoT Devices without coding.   GUIDE CONCEPT   Learn how to download, install, and configure the Edge Microserver (EMS) to create an AlwaysOn (TM) connection between Edge IoT Devices and ThingWorx Foundation.       YOU'LL LEARN HOW TO   Install the Edge MicroServer (EMS) Configure the EMS Connect the EMS to ThingWorx Foundation   NOTE: The estimated time to complete all parts of this guide is 30 minutes.     Step 1: Description   The Web Socket Edge MicroServer (WSEMS... or just EMS for short) is a pre-compiled application based on the C SDK.     Typically, the EMS is used on devices "smart" enough to have their own operating system, such as a Raspberry Pi or personal computer.   Rather than editing code and compiling into a custom binary (as with the SDKs), the EMS allows you to simply edit some configuration files to point the Edge IoT device towards the appropriate ThingWorx Foundation instance.   In addition, the EMS utilizes the PTC-proprietary AlwaysOn protocol to "phone-home", rather than having Foundation reach out to it. As such, the EMS will typically not require port forwarding/opening, and can easily communicate from a more-secure Edge environment to the Foundation server.     Step 2: Install EMS   In this step, you'll download and extract the EMS onto your personal Windows computer.   Versions of the EMS are available for Linux running on both x86 and ARM processors. Those are outside the scope of this guide, but require only minor modifications versus the instructions presented here.   Download the EMS.   Navigate to the directory where you downloaded the .zip file.     Extract the .zip file and explore into the extracted folder.     Navigate into and Copy all contents inside the \microserver directory.     Navigate to the C:\ root directory.     Create a C:\CDEMO_EMS folder. Note that this directory is not mandatory, but will be used throughout the rest of this guide.      Paste the contents of the extracted \microserver directory into C:\CDEMO_EMS.       Create Additional Directories   New folders may be added to the \CDEMO_EMS directory for various purposes.   Some of these will be utilized within this guide, while others may be utilized in future guides using the EMS.   Note that these particular names are not mandatory, and are simply the names used within this guide.    Create a C:\CDEMO_EMS\other directory. Create a C:\CDEMO_EMS\tw directory. Create a C:\CDEMO_EMS\updates directory.         Create Test Files   It can also be helpful during testing to have some small files in these folders to further demonstrate connectivity.   As these files were custom-created for the guide, seeing them within ThingWorx Foundation ensures that the connection between Foundation and the EMS is real and current.   In the C:\CDEMO_EMS\tw directory, create a text file named tw_test_01.txt. In the C:\CDEMO_EMS\other directory, create a text file named other_test_01.txt.     Click here to view Part 2 of this guide.

The Protocol Adapter Toolkit (PAT) is an SDK that allows developers to write a custom Connector that enables edge devices to connect to and communicate with the ThingWorx Platform.   In this blog, I will be dabbling with the MQTT Sample Project that uses the MQTT Channel introduced in PAT 1.1.2.   Preamble All the PAT sample projects are documented in detail in their respective README.md. This post is an illustrated Walk-thru for the MQTT Sample project, please review its README.md for in depth information. More reading in Protocol Adapter Toolkit (PAT) overview PAT 1.1.2 is supported with ThingWorx Platform 8.0 and 8.1 - not fully supported with 8.2 yet.   MQTT Connector features The MQTT Sample project provides a Codec implementation that service MQTT requests and a command line MQTT client to test the Connector. The sample MQTT Codec handles Edge initiated requests read a property from the ThingWorx Platform write a property to the ThingWorx Platform execute a service on the ThingWorx Platform send an event to the ThingWorx Platform (also uses a ServiceEntityNameMapper to map an edgeId to an entityName) All these actions require a security token that will be validated by a Platform service via a InvokeServiceAuthenticator.   This Codec also handles Platform initiated requests (egress message) write a property to the Edge device execute a service without response on the Edge device  Components and Terminology       Mqtt Messages originated from the Edge Device (inbound) are published to the sample MQTT topic Mqtt Messages originated from the Connector (outbound) are published to the mqtt/outbound MQTT topic   Codec A pluggable component that interprets Edge Messages and converts them to ThingWorx Platform Messages to enable interoperability between an Edge Device and the ThingWorx Platform. Connector A running instance of a Protocol Adapter created using the Protocol Adapter Toolkit. Edge Device The device that exists external to the Connector which may produce and/or consume Edge Messages. (mqtt) Edge Message A data structure used for communication defined by the Edge Protocol.  An Edge Message may include routing information for the Channel and a payload for Codec. Edge Messages originate from the Edge Device (inbound) as well as the Codec (outbound). (mqtt) Channel The specific mechanism used to transmit Edge Messages to and from Edge Devices. The Protocol Adapter Toolkit currently includes support for HTTP, WebSocket, MQTT, and custom Channels. A Channel takes the data off of the network and produces an Edge Message for consumption by the Codec and takes Edge Messages produced by the Codec and places the message payload data back onto the network. Platform Connection The connection layer between a Connector and ThingWorx core Platform Message The abstract representation of a message destined for and coming from the ThingWorx Platform (e.g. WriteProperty, InvokeService). Platform Messages are produced by the Codec for incoming messages and provided to the Codec for outgoing messages/responses.   Installation and Build  Protocol Adapter Toolkit installation The media is available from PTC Software Downloads : ThingWorx Connection Server > Release 8.2 > ThingWorx Protocol Adapter Toolkit Just unzip the media on your filesystem, there is no installer The MQTT Sample Project is available in <protocol-adapter-toolkit>\samples\mqtt Eclipse Project setup Prerequisite : Eclipse IDE (I'm using Neon.3 release) Eclipse Gradle extension - for example the Gradle IDE Pack available in the Eclipse Marketplace Import the MQTT Project : File > Import > Gradle (STS) > Gradle (STS) Project Browser to <protocol-adapter-toolkit>\samples\mqtt, then [Build Model] and select the mqtt project     Review the sample MQTT codec and test client Connector : mqtt > src/main/java > com.thingworx.connector.sdk.samples.codec.MqttSampleCodec decode : converts an MqttEdgeMessage to a PlatformRequest encode (3 flavors) : converts a PlatformMessage or an InfoTable or a Throwable to a MqttEdgeMessage Note that most of the conversion logic is common to all sample projects (websocket, rest, mqtt) and is done in an helper class : SampleProtocol The SampleProtocol sources are available in the <protocol-adapter-toolkit>\samples\connector-api-sample-protocol project - it can be imported in eclipse the same way as the mqtt. SampleTokenAuthenticator and SampleEntityNameMapper are also defined in the <protocol-adapter-toolkit>\samples\connector-api-sample-protocol project. Client : mqtt > src/client/java > com.thingworx.connector.sdk.samples.MqttClient Command Line MQTT client based on Eclipse Paho that allows to test edge initiated and platform initiated requests. Build the sample MQTT Connector and test client Select the mqtt project then RMB > Gradle (STS) > Task Quick Launcher > type Clean build +  [enter] This creates a distributable archive (zip+tar) in <protocol-adapter-toolkit>\samples\mqtt\build\distributions that packages the sample mqtt connector, some startup scripts, an xml with sample entities to import on the platform and a sample connector.conf. Note that I will test the connector and the client directly from Eclipse, and will not use this package. Runtime configuration and setup MQTT broker I'm just using a Mosquitto broker Docker image from Docker Hub​   docker run -d -p 1883:1883 --name mqtt ncarlier/mqtt  ThingWorx Platform appKey and ConnectionServicesExtension From the ThingWorx Composer : Create an Application Key for your Connector (remember to increase the expiration date - to make it simple I bind it to Administrator) Import the ConnectionServicesExtension-x.y.z.zip and pat-extension-x.y.z.zip extensions available in <protocol-adapter-toolkit>\requiredExtensions  Connector configuration Edit <protocol-adapter-toolkit>\samples\mqtt\src\main\dist\connector.conf Update the highlighted entries below to match your configuration :   include "application" cx-server {   connector {     active-channel = "mqtt"     bind-on-first-communication = true     channel.mqtt {       broker-urls = [ "tcp://localhost:1883" ]       // at least one subscription must be defined       subscriptions {        "sample": [ "com.thingworx.connector.sdk.samples.codec.MqttSampleCodec", 1 ]       }       outbound-codec-class = "com.thingworx.connector.sdk.samples.codec.MqttSampleCodec"     }   }   transport.websockets {     app-key = "00000000-0000-0000-0000-000000000000"     platforms = "wss://thingWorxServer:8443/Thingworx/WS"   }   // Health check service default port (9009) was in used on my machine. Added the following block to change it.   health-check {      port = 9010   } }  Start the Connector Run the Connector directly from Eclipse using the Gradle Task RMB > Run As ... > Gradle (STS) Build (Alternate technique)  Debug as Java Application from Eclipse Select the mqtt project, then Run > Debug Configurations .... Name : mqtt-connector Main class:  com.thingworx.connectionserver.ConnectionServer On the argument tab add a VM argument : -Dconfig.file=<protocol-adapter-toolkit>\samples\mqtt\src\main\dist\connector.conf Select [Debug]  Verify connection to the Platform From the ThingWorx Composer, Monitoring > Connection Servers Verify that a Connection Server with name protocol-adapter-cxserver-<uuid> is listed  Testing  Import the ThingWorx Platform sample Things From the ThingWorx Composer Import/Export > From File : <protocol-adapter-toolkit>\samples\mqtt\src\main\dist\SampleEntities.xml Verify that WeatherThing, EntityNameConverter and EdgeTokenAuthenticator have been imported. WeatherThing : RemoteThing that is used to test our Connector EdgeTokenAuthenticator : holds a sample service (ValidateToken) used to validate the security token provided by the Edge device EntityNameConverter : holds a sample service (GetEntityName) used to map an edgeId to an entityName  Start the test MQTT client I will run the test client directly from Eclipse Select the mqtt project, then Run > Run Configurations .... Name : mqtt-client Main class:  com.thingworx.connector.sdk.samples.MqttClient On the argument tab add a Program argument : tcp://<mqtt_broker_host>:1883 Select [Run] Type the client commands in the Eclipse Console  Test Edge initiated requests     Read a property from the ThingWorx Platform In the MQTT client console enter : readProperty WeatherThing temp   Sending message: {"propertyName":"temp","requestId":1,"authToken":"token1234","action":"readProperty","deviceId":"WeatherThing"} to topic: sample Received message: {"temp":56.3,"requestId":1} from topic: mqtt/outbound Notes : An authToken is sent with the request, it is validated by a platform service using the SampleTokenAuthenticator (this authenticator is common to all the PAT samples and is defined in <protocol-adapter-toolkit>\samples\connector-api-sample-protocol) EntityNameMapper is not used by readProperty (no special reason for that) The PlatformRequest message built by the codec is ReadPropertyMessage   Write a property to the ThingWorx Platform In the MQTT client console enter : writeProperty WeatherThing temp 20   Sending message: {"temp":"20","propertyName":"temp","requestId":2,"authToken":"token1234","action":"writeProperty","deviceId":"WeatherThing"} to topic: sample Notes : An authToken is sent with the request, it is validated by a platform service using the SampleTokenAuthenticator EntityNameMapper is not used by writeProperty The PlatformRequest message built by the codec is WritePropertyMessage No Edge message is sent back to the device   Send an event to the ThingWorx Platform   In the MQTT client console enter : fireEvent Weather WeatherEvent SomeDescription   Sending message: {"requestId":5,"authToken":"token1234","action":"fireEvent","eventName":"WeatherEvent","message":"Some description","deviceId":"Weather"} to topic: sample Notes : An authToken is sent with the request, it is validated by a platform service using the SampleTokenAuthenticator fireEvent uses a EntityNameMapper (SampleEntityNameMapper) to map the deviceId (Weather) to a Thing name (WeatherThing), the mapping is done by a platform service The PlatformRequest message built by the codec is FireEventMessage No Edge message is sent back to the device   Execute a service on the ThingWorx Platform ... can be tested with the GetAverageTemperature on WeatherThing ... Test Platform initiated requests     Write a property to the Edge device The MQTT Connector must be configured to bind the Thing with the Platform when the first message is received for the Thing. This was done by setting the bind-on-first-communication=true in connector.conf When a Thing is bound, the remote egress messages will be forwarded to the Connector The Edge initiated requests above should have done the binding, but if the Connector was restarted since, just bind again with : readProperty WeatherThing isConnected From the ThingWorx composer update the temp property value on WeatherThing to 30 An egress message is logged in the MQTT client console :   Received message: {"egressMessages":[{"propertyName":"temp","propertyValue":30,"type":"PROPERTY"}]} from topic: mqtt/outbound   Execute a service on the ThingWorx Platform ... can be tested with the SetNtpService on WeatherThing ...

Protocol Adapter Toolkit (PAT) is an SDK that allows developers to write a custom Connector that enables edge devices (without native AlwaysOn support) to connect to and communicate with the ThingWorx Platform. A typical use case is edge communication using a protocol that can't be changed (e.g. MQTT). Prior to PAT, developers had to use the ThingWorx (Edge or Platform) SDKs, or the ThingWorx REST interface, to enable the edge devices to communicate with ThingWorx. Overview PAT provides three main components: the Channel, the Codec, and the ThingWorx Platform Connection. The Channel implements a network protocol to communicate directly with the Edge Device. Its responsibilities include reading data from an Edge Device, writing data to an Edge Device, and routing data to the correct Codec. You can implement your own custom channel or use one of the out of the box channels provided by PTC : WebSocket, HTTP (1.0.x) and MQTT (1.1.x). The Codec translates messages from your edge devices into messages that ThingWorx platform can process (property read/write,service call, events), and provides a means to take the results of those actions and turn them back into messages for the device.  You must implement the Codec. The Platform Connection layer sends and receives messages with the ThingWorx platform. Note : The PAT Connector capabilities depend on edge protocol and channel implementation. Installation The PAT installation media contains : README.md - start here SDK (Java API) and runtime libraries PAT skeleton project (Gradle) Sample codec implementations for the WebSocket, HTTP, and MQTT channels (Gradle) Sample Custom Channel implementation (basic TCP protocol adapter) (Gradle) Required extensions to be installed on the platform : ConnectionServicesExtension and pat-extension Reference Documents ThingWorx Protocol Adapter Toolkit Developers Guide 1.0.0 README.md in various levels of installation folders ThingWorx Connection Services and Compatibility Matrix 1.0.0 Related Knowledge Protocol Adapter Toolkit - MQTT Sample Project hands-on (1.1.x)

Original Post Date:            September 30, 2016   Description: This tutorial video will walk you through the installation process for the PostgreSQL-based version of the ThingWorx Platform (7.2) in a RHEL environment.  All required software components will be covered in this video.    

Video Author:                     Stefan Tatka Original Post Date:            June 6, 2016   Description: This ThingWorx Tutorial will demonstrate how to configure and initiate remote file transfers using the .NET SDK.      

In the 8.2 release, we have upgraded our Mashup Runtime to jQuery 3.2.  This will give the platform a much needed upgrade to its core visualization library which will bring bug fixes, better performance, security enhancements, HTML5 compatibility and support for current browsers.  We will continue to upgrade all of our libraries across the platform with each ThingWorx release to ensure we are current and optimized.   We are releasing this functionality in 8.2 as an early preview and to enable regression testing of your ThingWorx applications.  In the Next Gen Composer, simply click on User Preferences and look for this setting:     Enabling jQuery3 runtime   JQuery 3 does introduce some new breaking changes (not from PTC!) that may affect existing apps.  We recommend turning on the JQuery3 option and testing your apps as soon as possible so there is time to fix any issues.  Please let PTC know if you are finding issues through our support site and our support staff will coach you through the upgrade process.  In 8.3, the jQuery 3 library will be our default for the Mashup design and runtime.  This means you will need to address any compatibly issues that jQuery 3 introduces (if any) to your widgets/applications before upgrading to the 8.4 release, where jQuery 3 will be the only available option.  We are hoping this dual mode, early access will help everyone through the transition and produce the best IoT applications possible!   You can also use the guides here for your reference: https://jquery.com/upgrade-guide/3.0/ https://github.com/jquery/jquery-migrate#migrate-older-jquery-code-to-jquery-30 https://blog.jquery.com/2017/03/16/jquery-3-2-0-is-out/

I have put together a small sample of how to get property values from a Windows Powershell command into Thingworx through an agent using the Java SDK. In order to use this you need to import entities from ExampleExport.xml and then run SteamSensorClient.java passing in the parameters shown in run-configuration.txt (URL, port and AppKey must be adapted for your system). ExampleExport.xml is a sample file distributed with the Java SDK which I’ve also included in the zipfile attached to this post. You need to go in Thingworx Composer to Import/Export … Import from File … Entities … Single File … Choose File … Import. Further instructions / details are given in this short video: Video Link : 2181

The accuracy of a predictive model can be boosted in two ways: Either by embracing Feature engineering or by applying boosting algorithms straight away. There are multiple boosting algorithms like Gradient Boosting, XGBoost, AdaBoost, Gentle Boost etc. Every algorithm has its own underlying mathematics and a slight variation is observed while applying them. While working with boosting algorithms, we have come across two frequently occurring buzzwords: Bagging and Boosting. Bagging: It is an approach where you take random samples of data, build learning algorithms and take simple means to find bagging probabilities. Boosting: Boosting is similar, however the selection of sample is made more intelligently. We subsequently give more and more weight to hard to classify observations. Below are Default Algorithms used in Predictive Models generated in ThingWorx Analytics: Decision Tree Gradient Boost Linear regression Neural Net Random Forrest Logistic Regression Gradient boosting is a machine learning technique for regression and classification problems, which produces a prediction model in the form of an ensemble of weak prediction models, typically decision trees. It builds the model in a stage-wise fashion like other boosting methods do, and it generalizes them by allowing optimization of an arbitrary differential loss function. Let’s begin with an easy example: Assume, you are given a previous model M to improve on. Currently you observe that the model has an accuracy of 80% (any metric). How do you go further about it? One simple way is to build an entirely different model using new set of input variables and trying better ensemble learners. On the contrary, we have a much simpler way to suggest. It goes like this: Y = M(x) + error What if we are able to see that error is not a white noise but have same correlation with outcome(Y) value. What if we can develop a model on this error term? Like:error = G(x) + error2 Probably, we will see error rate will improve to a higher number, say 84%. Let’s take another step and regress against error2: error2 = H(x) + error3 Now we combine all these together: Y = M(x) + G(x) + H(x) + error3 This probably will have a accuracy of even more than 84%. What if we can find an optimal weights for each of the three learners: Y = alpha * M(x) + beta * G(x) + gamma * H(x) + error4 How Gradient Boosting Works: 1. Loss Function: The loss function used depends on the type of problem being solved. It must be differential, but many standard loss functions are supported and you can define your own. A benefit of the gradient boosting framework is that a new boosting algorithm does not have to be derived for each loss function that may want to be used, instead, it is a generic enough framework that any differential loss function can be used. 2. Weak Learner: Decision trees are used as the weak learner in gradient boosting. Specifically regression trees are used that output real values for splits and whose output can be added together, allowing subsequent models outputs to be added and “correct” the residuals in the predictions. Trees are constructed in a greedy manner, choosing the best split points based on purity scores like Gini or to minimize the loss. 3. Additive Model: Trees are added one at a time, and existing trees in the model are not changed. A gradient descent procedure is used to minimize the loss when adding trees. we have weak learner sub-models or more specifically decision trees. After calculating the loss, to perform the gradient descent procedure, we must add a tree to the model that reduces the loss. Improvements to Basic Gradient Boosting: 1. Tree Constraints: It is important that the weak learners have skill but remain weak. Below are some constraints that can be imposed on the construction of decision trees: Number of trees: ​Generally adding more trees to the model can be very slow to over fit. The advice is to keep adding trees until no further improvement is observed. Tree depth: Deeper trees are more complex trees and shorter trees are preferred. Generally, better results are seen with 4-8 levels. Number of nodes or number of leaves: like depth, this can constrain the size of the tree, but is not constrained to a symmetrical structure if other constraints are used. Number of observations per split: Imposes a minimum constraint on the amount of training data at a training node before a split can be considered Minimum improvement to loss: Is a constraint on the improvement of any split added to a tree. 2. Weighted Updates: The contribution of each tree to this sum can be weighted to slow down the learning by the algorithm. This weighting is called a shrinkage or a learning rate. "Each update is simply scaled by the value of the “learning rate parameter v". 3. Stochastic Gradient Boosting: At each iteration a sub sample of the training data is drawn at random (without replacement) from the full training data set. The randomly selected sub sample is then used, instead of the full sample, to fit the base learner. 4. Penalized Gradient Boosting: The additional regularization term helps to smooth the final learnt weights to avoid over-fitting. Intuitively, the regularized objective will tend to select a model employing simple and predictive functions.

With the new licensing introduction, it could get confusing at first on how to obtain and apply, especially with more than one app in place. This is an example on how to apply both foundation and manufacturing license when installing Thingworx 8. 1) Install Manufacturing App 8.0 and needed components (ex: Kepware) per  the guide with manufacturing app license - manufacturing app widget can now be accessed. 2) Accessing /Thingworx reports a licensing issue 3) Download Thingworx license from the license portal. 4) Rename the manufacturing app license.bin to <name>.bin and put Thingworx license.bin in the ThingworxPlatform folder. 5) Restart Thingworx service 6) Access /Thingworx and accept license agreement 7) Change license.bin back to the original manufacturing app license.bin (step 4) 😎 Restart Thingworx server 9) Both manufacturing app and foundation functions are available.

The KEPServerEX ThingWorx Native Interface was originally released with KEPServerEX v5.21 to enable connectivity with v6.6 of the ThingWorx Platform. Compatibility with the ThingWorx NextGen Composer (v7.4) was implemented with the release of KEPServerEX v6.1, while still allowing support of older ThingWorx composer versions through the Native Interface "Legacy mode" setting.   When connecting with ThingWorx v7.4 and newer, an Industrial Gateway thing template is configured in the NextGen Composer. The instructions for connecting KEPServerEX v6.1 (and newer) with ThingWorx v7.4 (and newer) can be found in the ThingWorx Help Center here:   KEPServerEX / ThingWorx Industrial Connectivity Connection Example (Expand ThingWorx Model Definition and Composer > Industrial Connections > Industrial Connections Example)     When connecting with versions of ThingWorx that are older than v7.4-- or if the older version of ThingWorx Composer will be used-- it will necessary to download and import the KEPServerEX Extension from the ThingWorx marketplace. The "RemoteKEPServerEXThing" thing template will then be used to allow connection with KEPServerEX. Here is a link to the extension download: ThingWorx IoT Marketplace   To enable Legacy Mode in KEPServerEX (v6.1 or newer, when connecting with versions of ThingWorx that are older than v7.4): 1. Open the KEPServerEX Configuration window 2. In the top-left Project view pane, right-click on "Project" and select "Properties" 3. Select the ThingWorx Property Group 4. Change "Enable" to "Yes"; and change "Legacy Mode" to "Enable"     See the chart below for a version compatibility summary:   KEPServerEX version ThingWorx version v5.21 pre-7.4, using RemoteKEPServerEXThing v6.0 pre-7.4, using RemoteKEPServerEXThing v6.1 or higher w/ Legacy Mode enabled pre-7.4, using RemoteKEPServerEXThing v6.1 or higher w/ Legacy Mode disabled (default) v7.4 or higher, using Industrial Gateway

When I tried to set String property values with Chinese letters by using C SDK, I could see only broken characters in Thingworx. The cause of the problem is simple and it's encoding. In order to solve this problem, you need to convert encoding to UTF-8 in your C code. I used 'libiconv' library to do it. This guide is for Win32 version. If you want to make a library for other platforms such as Linux, you can create or get a library for your own by googling. How to Get the Source Code of libiconv At the moment, the most recent version of libiconv is 1.15. You can download the source code of libiconv from here. How to Build I used MS Visual Studio 2012, but the explanation can be applied to the earlier versions of MS Visual Studio and express editions. Step 1. Download the most recent version of libiconv and unzip the file. Step 2. Make a new Win32 Project. Let's say "libiconv" as the project name. Check to create directory for solution. Choose DLL as the application type and check Empty Project for additional options. Click the button "Finish" to generate the new project. Step 3. Copy files from the folders of libiconv to project folders. To build "libiconv", you need to compile three files "localcharset.c", "relocatable.c" and "iconv.c". That's the key! Copy three files "relocatable.h", "relocatable.c" and "iconv.c" in the folder "...\libiconv-1.15\lib\" to the project folder "...\libiconv\libiconv\". Copy "...\libiconv-1.15\libcharset\lib\localcharset.c" to the project folder "...\libiconv\libiconv\". Copy "...\libiconv-1.15\libcharset\include\localcharset.h.build.in" to the project folder "...\libiconv\libiconv\" and then, rename the copied "localcharset.h.build.in" to "localcharset.h". Copy "...\libiconv-1.15\windows\libiconv.rc" to the project folder "...\libiconv\libiconv\". Make folder "include" under the project folder "...\libiconv\". Copy "...\libiconv-1.15\include\iconv.h.build.in" to the project include folder "...\libiconv\include" and then, rename the copied "iconv.h.build.in" to "iconv.h". Copy "...\libiconv-1.15\config.h.in" to the project include folder "...\libiconv\include" and then, rename the copied "config.h.in" to "config.h". Copy all the header files (*.h) and definition files (*.def) in the folder "...\libiconv-1.15\lib" to the project include folder "...\libiconv\include". Step 4. Add existing items. Execute "project > Add Existing items..." at the main menu to add existing items to the project. Step 5. Project Settings. You can make 64-bit platform through configuration manager in order to generate libiconv.dll for 64-bit system. You can also make two other configurations "ReleaseStatic" and "DebugStatic" in order to generate libiconvStatic.lib as a static link library. At the project properties, change Output Directory as "$(SolutionDir)$(Configuration)_$(Platform)\" and Intermediate Directory as "$(SolutionDir)obj\$(ProjectName)\$(Configuration)_$(Platform)\". Change Include Directories as "..\include;$(IncludePath)": You have to add "BUILDING_LIBICONV" and "BUILDING_LIBCHARSET" to Peprocessor Definitions of all Platforms and of all configurations. You'd better set Runtime Library to "Multi-threaded" when building dynamic link library libiconv.dll. Then, the dependency on VC Runtime library can be controlled by the applications that will be built and dynamically linked with libiconv.dll because libiconv.dll does not need VC Runtime library but only the application that uses libiconv.dll may or may not need VC Runtime library. However, when building the static link library libiconvStatic.lib, you can choose Runtime Library option for libiconvStatic.lib depending on the application that uses libiconvStatic.lib. You have to change Precompiled Header option to "Not Using Precompiled Headers". Step 6. Tweak the source code. libiconv.rc Open libiconv.rc with text editor or the source code editor of Visual Studio IDE by double-clicking libiconv.rc in the Solution explorer and insert some code at line 4 as follows: /////////////    ADD    ///////////// #define PACKAGE_VERSION_MAJOR 1 #define PACKAGE_VERSION_MINOR 14 #define PACKAGE_VERSION_SUBMINOR 0 #define PACKAGE_VERSION_STRING "1.14" ///////////////////////////////////// You may be asked to change Line endings to "Windows (CR LF)". Then, let it do so. It will be more convenient for you if you mainly use Windows. localcharset.c Open localcharset.c and delete or comment the lines 80 - 83 as follows: //////////////////  DELETE //////////////// ///* Get LIBDIR.  */ //#ifndef LIBDIR //# include "configmake.h" //#endif /////////////////////////////////////////// iconv.c Open iconv.c and delete or comment the lines 250 - 252 and add three lines there as follows: ///////////////////////// DELETE /////////////////////// //size_t iconv (iconv_t icd, //              ICONV_CONST char* * inbuf, size_t *inbytesleft, //              char* * outbuf, size_t *outbytesleft) /////////////////////////   ADD   ////////////////////// size_t iconv (iconv_t icd,               const char* * inbuf, size_t *inbytesleft,               char* * outbuf, size_t *outbytesleft) //////////////////////////////////////////////////////// localcharset.h Open localcharset.h and delete or comment the lines 21 - 25 and add 7 lines there as follows: /////////////////////////   DELETE  //////////////////////// //#if @HAVE_VISIBILITY@ && BUILDING_LIBCHARSET //#define LIBCHARSET_DLL_EXPORTED __attribute__((__visibility__("default"))) //#else //#define LIBCHARSET_DLL_EXPORTED //#endif /////////////////////////    ADD    ////////////////////// #ifdef BUILDING_LIBCHARSET #define LIBCHARSET_DLL_EXPORTED __declspec(dllexport) #elif USING_STATIC_LIBICONV #define LIBCHARSET_DLL_EXPORTED #else #define LIBCHARSET_DLL_EXPORTED __declspec(dllimport) #endif //////////////////////////////////////////////////////////////////// config.h Open config.h in the project include folder "...\libiconv\include" and delete or comment the lines 29 - 30 as follows: ///////////////////////// DELETE /////////////////////// ///* Define as good substitute value for EILSEQ. */ //#undef EILSEQ //////////////////////////////////////////////////////// Otherwise you can redefine EILSEQ as good substitute value. iconv.h Open iconv.h in the project include folder "...\libiconv\include" and delete or comment the line 175 and add 1 line as follows: /////////////////////////  DELETE  /////////////////////// //#if @HAVE_WCHAR_T@ /////////////////////////    ADD   ////////////////////// #if HAVE_WCHAR_T //////////////////////////////////////////////////////////////////////////////// Delete or comment the line 128 and add 1 line as follows: /////////////////////////  DELETE  /////////////////////// //#if @USE_MBSTATE_T@ /////////////////////////   ADD   ////////////////////// #if USE_MBSTATE_T //////////////////////////////////////////////////////////////////////////////// Delete or comment the lines 107-108 and add 2 lines as follows: /////////////////////////  DELETE  /////////////////////// //#if @USE_MBSTATE_T@ //#if @BROKEN_WCHAR_H@ /////////////////////////  ADD  ////////////////////// #if USE_MBSTATE_T #if BROKEN_WCHAR_H //////////////////////////////////////////////////////////////////////////////// Delete or comment the line 89 and add 2 lines as follows: /////////////////////////  DELETE /////////////////////// //extern LIBICONV_DLL_EXPORTED size_t iconv (iconv_t cd, @ICONV_CONST@ char* * inbuf, //size_t *inbytesleft, char* * outbuf, size_t *outbytesleft); /////////////////////////    ADD   ////////////////////// extern LIBICONV_DLL_EXPORTED size_t iconv (iconv_t cd, const char* * inbuf,   size_t *inbytesleft, char* * outbuf, size_t *outbytesleft); //////////////////////////////////////////////////////////////////////////////// Delete or comment the lines 25 - 30 and add 8 lines as follows: /////////////////////////  DELETE /////////////////////// //#if @HAVE_VISIBILITY@ && BUILDING_LIBICONV //#define LIBICONV_DLL_EXPORTED __attribute__((__visibility__("default"))) //#else //#define LIBICONV_DLL_EXPORTED //#endif //extern LIBICONV_DLL_EXPORTED @DLL_VARIABLE@ int _libiconv_version; /* Likewise */ /////////////////////////    ADD   ////////////////////// #if BUILDING_LIBICONV #define LIBICONV_DLL_EXPORTED __declspec(dllexport) #elif USING_STATIC_LIBICONV #define LIBICONV_DLL_EXPORTED #else #define LIBICONV_DLL_EXPORTED __declspec(dllimport) #endif extern LIBICONV_DLL_EXPORTED int _libiconv_version; /* Likewise */ //////////////////////////////////////////////////////////////////////////////// How to Use When you use newly built libiconv, the only header file that you need is iconv.h. You will need to link either the import library libiconv.lib or the static library libiconvStatic.lib in your project property or write the code in one of your source file as follows: #pragma comment (lib, "libiconv.lib") or #pragma comment (lib, "libiconvStatic.lib") In the source of the application that uses this library either libiconv.dll or libiconvStatic.lib, if you don't define anything but only include iconv.h, your application will use libiconv.dll while it will use libiconvStatic.lib if you define USING_STATIC_LIBICONV before you include iconv.h in your application as follows: //#define USING_STATIC_LIBICONV #include <iconv.h> And in C SDK code, I used a "SteamSensorWithFileTransferAndTunneling" sample code and added codes in the "dataCollectionTask" function as below. void dataCollectionTask(DATETIME now, void * params) {   iconv_t ic;   char* in_buf = "Hi, 文健英";   char *to_chrset = "UTF-8";   char *from_chrset = "EUC-KR";   size_t in_size = strlen(in_buf);   size_t  out_size = sizeof(wchar_t) * in_size * 4;   char* out_buf = malloc(out_size);   // Caution: iconv's inbuf, outbuf are double pointers, so need to define separate pointers and pass addresses.   char* in_ptr = in_buf;   char* out_ptr = out_buf;   size_t out_buf_left;   size_t result;   memset(out_buf, 0x00, out_size);      ic = iconv_open(to_chrset, from_chrset);   if (ic == (iconv_t) -1)   {         printf("Not supported code \n");         exit(1);   }   printf("input len = %d, %s\n", in_size, in_buf);   out_buf_left = out_size;   iconv(ic, &in_ptr, &in_size, &out_ptr, &out_buf_left);   //printf("input len = %d, output len=%d %s\n", in_size, out_size - out_buf_left, out_buf);   iconv_close(ic);   /* TW_LOG(TW_TRACE,"dataCollectionTask: Executing"); */   properties.TotalFlow = rand()/(RAND_MAX/10.0);   properties.Pressure = 18 + rand()/(RAND_MAX/5.0);   properties.Location.latitude = properties.Location.latitude + ((double)(rand() - RAND_MAX))/RAND_MAX/5;   properties.Location.longitude = properties.Location.longitude + ((double)(rand() - RAND_MAX))/RAND_MAX/5;   properties.Temperature  = 400 + rand()/(RAND_MAX/40);   properties.BigGiantString = out_buf; // Set values for String property   /* Check for a fault.  Only do something if we haven't already */   if (properties.Temperature > properties.TemperatureLimit && properties.FaultStatus == FALSE) {   twInfoTable * faultData = 0;   char msg[140];   properties.FaultStatus = TRUE;   properties.InletValve = TRUE;   sprintf(msg,"%s Temperature %2f exceeds threshold of %2f",   thingName, properties.Temperature, properties.TemperatureLimit);   faultData = twInfoTable_CreateFromString("message", msg, TRUE);   twApi_FireEvent(TW_THING, thingName, "SteamSensorFault", faultData, -1, TRUE);   twInfoTable_Delete(faultData);   }   /* Update the properties on the server */   sendPropertyUpdate(); }

This video is the 3 rd part of a series of 3 videos walking you through how to setup ThingWatcher for Anomaly Detection. In this video we will use Anomaly Mashup to visualize data received from my remote device.   Updated Link for access to this video:  Anomaly Detection 8.0:  Viewing Data via Anomaly Mashup:  Part 3 of 3

This video is the 2 nd part of a series of 3 videos walking you through how to setup ThingWatcher for Anomaly Detection. In this video you will learn how to use “Discover UI” from the “New Composer” to bind simulated data coming through KEPServer for Anomaly Detection.   Updated Link for access to this video:  Anomaly Detection 8.0: Configuring Anomaly Alerts:  Part 2 of 3

This blog post provide information on the technical changes in Thingworx 8.0, New Technical Changes in ThingWorx 8.0.0 Here are some common questions and answers in regards to the Licensing change: Does that mean all the extensions in the marketplace won't be free anymore? Depends on the extensions. The main extensions we are licensing for 8.0 are Navigate, Manufacturing and Utilties. We are not licensing the MailExtension on the marketplace, for example. Partners and customers can still import their custom apps/extensions If TWX connects to RP(remote platform) which has its own subscription based Flexera license (InService, for example), how does this interaction works- license validation.  Is server to sever connection counts as user login direct to PTC product? License files are per TWX instance. For RP, each would have their own license files. User counts (if entitled and enforced) are generic to each system.

  You might have seen the Performance Advisor for some of your other favorite PTC Products like Creo, Windchill or Integrity.  Good news....it's now also available for ThingWorx!   In case you're not familiar with the Performance Advisor, it's new functionality allowing you to work closer with the PTC / ThingWorx team for improving your usage with ThingWorx and improving ThingWorx itself in the areas that matter most to you.      ThingWorx Performance Advisor   delivers information dashboards driven by data on the features, usage and performance of your ThingWorx systems unlocks information that can reduce wasted development and improve design cycles allows comprehensive visibility into software versions in use to manage software upgrade plans simplifies compliance and revenue allocation by monitoring usage enables quick access to system and usage statistics across your organization uses personalized dashboards to viewing, reporting and trend analysis   The Performance Advisor for ThingWorx has just been released, so we want you to share your experience and data to get you and us started on analyzing usage statistics and needs for further features.   The Performance Advisor is easy to connect. It just takes three simple steps and a minute of your time. This will result in improved transparency, improved stability, improved productivity, improved product performance, improved compliance administration and an increased administrative efficiency and allows the ThingWorx R&D team to continuously improve the platform through the analytical insights from the data collected.   As ThingWorx is growing fast, be sure to participate and actively shape the way you're using ThingWorx and the way that ThingWorx is designed.   With newer versions of ThingWorx, capabilites and benefits for the Performance Advisor will be improved to ensure we're capturing the most accurate information to help you grow your Internet of Things business and scale your solutions to your / your application's needs and requirements. We're just at the beginning of the journey...   How to enable ThingWorx Performance Advisor   Enable Metrics Reporting and setting up the Performance Advisor capabilties is described in detail in CS262960 Just follow the steps and: Congratulations!   It's as simple and fast as that - you enabled the ThingWorx Performance Advisor... quite easy, right?   Where can I see the data / metrics I have sent to PTC?   The information can be seen on the Performance Advisor Homepage   Here's how the current views look like - they might change over time, introducing new features and views to maximize the impact and benefit for you.   In a first glance the basic information of what has been collected can be seen in the Summary     In the Connection System Details it shows more about what systems are currently connected with its user counts and number of remote things. The Connected System History shows a historical overview on how those parameters changed over time.   For a more detailed historic overview of all the data being sent, check out the Historical Property Data.     Questions?   For specific questions, check out article CS262967 which holds the FAQs for the Performance Advisor   If you have specific questions not addressed in the article, you can always comment on this blog post, open a new community thread or open a case with Support Services.   We want your feedback   After enabling metrics collection and reviewing the Performance Advisor dashboards, what do you think? What features would you like to see in the future? Is there anything missing that would help you as a System Administrator making your life easier?   As we're trying to improve functionality over time, make sure your voice is heard as well and feel free to leave some feedback.

All of the entities and data in storage entities(Stream, ValueStream, DataTable, Wiki, Blog) are kept in Persistence Provider, to enhance the storage volume of Thingworx, it is advised to use bigger Persistence Provider (PostgreSQL or Cassandra rather than Neo4j and H2), move and keep certain part of data in external database or set up a second Persistence Provider Data Storage entities (Stream, ValueStream, DataTable) and Collaboration entities (Wiki, Blog) can be assigned to the new Persistence Provider to keep the Business Data The original Persistence Provider (upon installation) will be freed from large data input and respond faster, and only holds the Model Data (entity information) Small Persistence Provider like H2 or Neo4j, which does not require extra steps to setup in the Thingworx installation process, cannot set up a second Persistence Provider (they don't have .bat files to configure the database) Here is an example for setting up a second PostgreSQL (may also applicable for Cassandra) Install a new PostgreSQL in the server Should give different folder names for the files, or two PostgreSQL will affect each other PostgreSQL 9.5 could work with Thingworx 7.2.x, but cannot be used with 7.3 and above. Version 9.5 is not officially supported or advised by now, and the user should take their own risk if doing so. The new PostgreSQL will have a new port number (e.g. first port is 5432, the new port is 5433) Open PostgreSQL using pgAdmin 3 Create a new user role: a. Right click PostgreSQL9.4 (localhost:5433). b. Select NewObject>New Login Role. On the Properties tab, in the Role name field, enter the user role name. c. On the Definition tab, in the Password field, enter a unique password (you will be prompted to enter it twice). Add the new <postgres-installation>/bin folder to system path variable Edit the thingworxPostgresDBSetup.bat and thingworxPostgresSchemaSetup.bat from the Thingworx software download package, and change the port property to 5433 Execute the two scripts, and new database and tablesapce will be created platform-settings.json file does not need extra configuration Restart the Tomcat server, a new folder will be created under ThingworxPostgresqlStorage folder Go to Thingworx composer, create a new Persistence Provider in the Home list Give it a new name, and select Persistence Provider Package(only one choice) After selection, a Configuration field will show up in the ENTITY INFORMATION field ( such field will not appear in Neo4j based Thingworx) Open Configuration and change the JDBC URL to jdbc:postgresql://localhost:5433/thingworx, update Password, click Save The new Persistence Provider can be used in data storage entities now The new Persistence Provider can also be set as default provider