Jens Bertenbreiter, Jürgen Rother, Düsseldorf, März 2018

Size: px
Start display at page:

Download "Jens Bertenbreiter, Jürgen Rother, Düsseldorf, März 2018"

Transcription

1 Jens Bertenbreiter, Jürgen Rother, Düsseldorf, März 2018 BASEL BERN BRUGG DÜSSELDORF FRANKFURT A.M. FREIBURG I.BR. GENF HAMBURG KOPENHAGEN LAUSANNE MÜNCHEN STUTTGART WIEN ZÜRICH

2 Use cases

3 A choice 2016 / 2017 IoT / SmartMeter Data Lake / Architecture Data Science Orientation Governance Training Technology Evaluation Infrastructure MeineStadt Big- and Fast Data Reference Projects 3

4 Big Data Customer Cases (excerpt) Big- and Fast Data Reference Projects 4

5 IWB Industrielle Werke Basel Smart Asset Management Objectives IWB owns many energy producing assets all around Europe and Switzerland. The objective was to connect all assets to the Azure Cloud to gain real time insights into the facilities to improve efficiency and reduce downtime. Tactics 144 wind turbines, four district heating power plant and two hydroelectric power statios were connected to the Azure Cloud. All sensordata is updated with masterdata, stored in a central database and provided to the business user in datamarts. Results Reduced maintenance and down time resulted in an increased revenue of serveral 100k CHF per year. Ability to use predictive maintenance for further efficiency improvements Full transparency on their energy producing assets and the local operators "One of the major challenges in the energy sector is how to collect data and make it available for further usage. The deployment of IoT technology makes it possible to develop customer-driven and near-time solutions that are cost-effective, powerful, and expandable." René Frei, Head of ICT, IWB.

6 Transnet BW Process data archive SCADA DWH Objectives TransnetBW is responsible for the operation, maintenance, planning and demand-driven expansion of the transmission network in Baden-Württemberg. The volume of data to be analyzed significantly increased. This was hardly feasible with the existing systems, which were not developed for this task in their basic structure. Tactics A new process data archive was build. In it, all prioritized data is collected in a central location, archived and made available for analytics. The data includes measurement and calculation which provide detailed information about the status of the system. Results Archiving 65 million records daily Over 71 billion records since 2014 Modern, scalable platform for future requirments "Our main objective is to ensure that the electricity grid is operated safely and to ensure the balance of production and consumption in the network at all times. To do this, we need to monitor and control the grid completely. We support the jointly developed process data archive." Matthias Wolf, data manager, TransnetBW

7 Industry Simulator Cloud re-player for complex industrial facilities Objectives Changing large industrial plants is complex, complicated and of high risk. Use cases of the departments must therefore be tested extensively before they can be integrated into the plants. However, digitalization requires a quick response to the needs and rapid testing of ideas. Tactics In Azure, a simulator has been set up in which existing data can be loaded. With this data, adjustments can be tested and the results analyzed and the use case can be adopted until it perfectly fits the needs. Results Quick and easy prove of use cases Trying ideas fast and flexible Keeping the industrial facilities stable

8 Big Data architecture blueprint

9 Trivadis Reference Architecture for Modern Data Analytic Solutions

10 Building Blocks M1 Batch Data Ingestion M2 Streaming Data Ingestion M3 Big Data Batch Processing M4 Event-/Stream-Processing M5 Analytics and Machine Learning M6 Accessing the Data Lake M7 Pushing Data from Data Lake to external systems M8 Managing the Data Lake M9 Master Data Management

11 Data Ingestion Batch ingestion Streaming ingestion

12 Data Processing Batch processing Stream processing

13 Analytics & Machine Learning Analytics & ML

14 Accessing Data Pushing Data to DWH Accessing Data Lake

15 Example end to end process Data Ingestion Data Processing Data source Collect data Save data to raw storage Format Translation Enrich with Metadata Save to redefined data storage Data Analytics Data Access Merge Data & Perform Analytics Save data to usage optimized data storage Map to Table & Access Control Access through SQL Visualize & Reporting

16 Fragen und Antworten Jens Bertenbreiter Tel.: Jürgen Rother Tel.: