Sofia2 IoT Platform release 4.0 now available!
This new version is available in our Open Platform in Sofia2 CloudLab.
View other releases
Following are the new features and improvements included in this new release:
GSMA DATA MODEL SUPPORT
GSMA association is working on an IoT Big Data Harmonised Data Model. GSMA Data Models (also called FIWARE DataModel) are defined in JSON, so its representation as a Sofia2 Ontology is straightforward.
Now you have Sofia2 templates that will allow you to directly work with these data models.
OPEN DATA CKAN PORTAL INTEGRATION
Now you can configure easily a bidirectional integration between Sofia2 and CKAN Open Data Portal. This feature is allowing the integration with the repositories of those Public Administration/Intitutions that are using this standard.
The platform allows to:
- Register a CKAN repository
- DataSets import from a CKAN repository
- APIs and Ontologies export as DataSets into a CKAN repository
At the time of registering a CKAN respository, we can do it in two ways, Reading mode: If we don’t have any user and API-KEY in order to publish into CKAN repository or Read & write mode: in case we have the required access rights (user an API-KEY)
Once the repository is registered, we can see the different datasets published on it.
, together with its details:
, including the option of visualizing the dataset in the different formats supported by the portal (usually JSON, XML and CSV).
You can import any DATASET in any of these formats as an ontology in SOFIA2,
Once imported, we will have all the capabilities of Sofia2 ontologies appied to the dataset imported: queries, visualization, analytics…
In case we want to publish an ontology into a remote CKAN, we will create an API using Sofia2 APIManager module, and we will publish it as Dataset with an URL linked to it. In this way, CKAN will work as Open Data Portal on the information managed by Sofia2 Platform.
ONTOLOGIES CREATION AND LOAD FROM JSON AND XML FILES.
In this release, we are improving the capabilities of the platform to create & load ontologies from external files. We are adding the new capability of doing so from XML and JSON Files, to the already known capability of doing this from CSV&Excel files.
You will find a new option on the screen for Ontologies Creation, with this new option. You will be able to select the file and load JSON/XML fields, and select the data type of each one of them.
As final step we will generate the schema and load the data in order to create the ontology.
TIME SERIES ONTOLOGIES SUPPORT:
This is a new type of ontology that will support the concept of time series. This new concept registers events together with its timestamp included in a time-slot bigger that groups them in a single document ( and stored in our Real Time Data base). This is optimizing the queries of this events, because with just one query we will get all of them or a subset of them in the same time-slot.
Ontology type Temporal Series can be created from the Ontologies Menu, where you can configure the time-slot, the grouping unit, the frequency of the events and the type of the data:
A Time Series ontology will look like this:
(The simple above is about an hour-slot time, grouping by minutes, and getting new data each second. In this way, we would have and ontology grouped in 60 groups of minutes all the 60 events reached every second.)
In terms if querying, we can for example launch the following SQL sentence:
Getting this resultset:
NEW NOTEBOOKS MODULE VERSION
Sofia2 Notebooks module is based on Zeppelin open-source software. In this release we are including the new version of this product and we are improving it with some new features:
We can highlight the following new functionalities supported:
- Export/Import of Notebooks in JSON files format.
- Export of the data generated in the visualizations in TSV/CSV format
- Notebooks versión control using different repositories:
- New visualization capabilities thanks to the incorporation of Helium framework.
- Updated to Zeppelin 0.7 and Spark 2.1 versions.
- Centralized visualization of the information related to notebooks execution (Jobs)
- Security improvements
SOFIA2 CLIENT COMPONENT FOR NODE-RED:
You now have available a set of customized Sofia2 Nodes, so you can use easilly SOFIA2 as backend from Node-RED
Using these nodes we will be able to do actions such as insert, update, delete, query and release a connection
NOTEBOOKS ORCHESTRATION FROM FLOWS ENGINE:
With this you will be able to orchestrate notebooks processes, even parallelizing the execution of the same notebook process and distributing the load.
You will be able also to use dynamic parameters (for input or output) at the moment of starting the notebooks process; including in this notebooks fixed information or information generated by another flow, using whatever language from the ones supported by Notebooks module, and initiate an execution with it afterwards:
With all this, we can create complex flows like this one below:
SAMPLE WEB APPLICATION USING ANGULARJS, SPRING BOOT AND SOFIA2 AS BACKEND:
AngularJS (at the fronted layer) and Spring boot (at the REST integration layer) are one of the most used technologies at the moment.
In this released we prepared a web application skeleton using it, so you can use it now as a template to build your web applications on Sofia2 as backend.
The simple represents a devices catalogue, including the actions of listing, creating a new device, show details and remove elements:
It is including also Sofia2 authentication and the use of the archetype_phone ontology.
You can find it in our github repository as a Maven project
DEPLOYMENT & PLATFORM OPERATION IMPROVEMENTS
Now we have a complete deployment using Docker and kubernetes, so deploying and increasing the elasticity is more simple now.
Also, following DevOps Principles, te lifecycle management of each reléase is now more automated (from development, Continuous Integration and deployment using a container orchestrator such as Kubernetes):
Most of you were waiting for this new functionality, that will allow you to apply filters to the gadgets configured in Sofia2 Dashboards.
In this way, I can configure with no coding at all, dashboards with this type of elements:
As you can see, you can create different type of filters: simple, multiple or in a date range.
You can apply this to any of the current gadgets, including the maps:
GADGETS & DASHBOARDS NEW FEATURES AND IMPROVEMENTS:
Or look&feel improvements in different panels configuration:
You also can now export data represented in a dashboard in CSV format, with the download link on each gadget:
MEETUPS IoT & BigData Sofia2 Lab GROUP CREATED
Meetups are virtual groups of people interested in common themes that meet to debate and expose about these themes (in our case, about technology and innovation).
The group has been created with the intention of holding meetings and events at least once a month. It is intended that these meetings are not only informative, but also involve attendees in the realization of workshops and practical examples around Sofia2 and IoT and Big Data technologies.
Next May 4 will be held the third meetup, which will test the analytical capabilities of the platform. (https://www.meetup.com/es-ES/IoT-BigData-Sofia2-Lab/events/238898432/)