Presentation is loading. Please wait.

Presentation is loading. Please wait.

Real-time Fraud DetectionStreaming ETLPredictive MaintenanceCall Center Analytics IT Infrastructure and Network Monitoring Customer behavior.

Similar presentations


Presentation on theme: "Real-time Fraud DetectionStreaming ETLPredictive MaintenanceCall Center Analytics IT Infrastructure and Network Monitoring Customer behavior."— Presentation transcript:

1

2

3

4

5

6

7 Real-time Fraud DetectionStreaming ETLPredictive MaintenanceCall Center Analytics IT Infrastructure and Network Monitoring Customer behavior predictionLog AnalyticsReal-time cross sell offers Fleet monitoring and Connected CarsReal-time Patient MonitoringSmart GridReal-time Marketing and many more…

8

9 Presentation and action Storage and Batch Analysis Stream Analytics Event Queuing & Stream Ingestion Event production IoT Hubs Applications Archiving for long term storage/ batch analytics Real time dashboard Stream processing Automation to kick-off workflows Machine Learning Reference Data Event Hubs Blobs Devices & Gateways

10 Development costs Infrastructure – procure & setup Cluster management & topology construction Integrations with ingress, egress & ML Resiliency and HA Scaling with business growth Monitoring & troubleshooting Operational costs Time

11 Maximum CONTRO L Greatest SIMPLICITY. Lowest TCO Azure Stream Analytics Fully managed – 3 9s of availability SQL like query language Machine Learning integrations Integrations for real time Dashboarding Custom code extensions Spark Streaming & Storm (HD Insight) SLA with 3 9s of availability Virtual Machines Full stack control Integrations with Kafka, ipython notebook R- Server Managed Spark Streaming & Storm

12

13 Storm on HDInsight Spark Streaming on HD Insight Azure Stream Analytics Strictest guaranteeAt-least-once (exactly once with Trident) Exactly onceAt least once Processing modelEvent at a timeMicro batching ScalingManual Fully automated State management Yes yes Programing modelJava, C#Scala, Python, JavaSQL like query language Open sourceYes No

14 Programmer productivity Ease of use Lowest TCO 3 lines of code in Azure Stream Analytics @ApplicationAnnotation(name="WordCountDemo") public class Application implements StreamingApplication { protected String fileName = "com/datatorrent/demos/wordcount/samplefile.txt"; private Locality locality = null; @Override public void populateDAG(DAG dag, Configuration conf) { locality = Locality.CONTAINER_LOCAL; WordCountInputOperator input = dag.addOperator("wordinput", new WordCountInputOperator()); input.setFileName(fileName); UniqueCounter wordCount = dag.addOperator("count", new UniqueCounter ()); dag.addStream("wordinput-count", input.outputPort, wordCount.data).setLocality(locality); ConsoleOutputOperator consoleOperator = dag.addOperator("console", new ConsoleOutputOperator()); dag.addStream("count-console",wordCount.count, consoleOperator.input); } @ApplicationAnnotation(name="WordCountDemo") public class Application implements StreamingApplication { protected String fileName = "com/datatorrent/demos/wordcount/samplefile.txt"; private Locality locality = null; @Override public void populateDAG(DAG dag, Configuration conf) { locality = Locality.CONTAINER_LOCAL; WordCountInputOperator input = dag.addOperator("wordinput", new WordCountInputOperator()); input.setFileName(fileName); UniqueCounter wordCount = dag.addOperator("count", new UniqueCounter ()); dag.addStream("wordinput-count", input.outputPort, wordCount.data).setLocality(locality); ConsoleOutputOperator consoleOperator = dag.addOperator("console", new ConsoleOutputOperator()); dag.addStream("count-console",wordCount.count, consoleOperator.input); }

15 Azure SQL DB Azure Event Hubs Azure Tables Azure Blob Storage Azure IoT Hub Azure Event Hubs Reference Data Query runs continuously against incoming streams of events Events Optionally, have a defined schema and are temporal (sequenced in time)

16 Temporal Semantics in ASA Application or ingest timestamp Windowing functions Policies for event ordering Policies to manage latencies between ingress sources Manage streams with multiple timelines Join multiple streams on temporal windows Join streaming data with data at rest

17 Data Manipulation SELECT FROM WHERE HAVING GROUP BY CASE WHEN THEN ELSE, INNER/LEFT OUTER JOIN, UNION, CROSS/OUTER APPLY, CAST, INTO, ORDER BY ASC, DSC Powerful analytics using simple SQL constructs Scaling Extensions WITH, PARTITION B Y OVER Date and Time Functions DateName, DatePart, Day Month, Year, DateDiff DateTimeFromParts, DateAdd Windowing Extensions TumblingWindow HoppingWindow SlidingWindow Aggregate Functions SUM, COUNT, AVG, MIN, MAX, STDEV, STDEVP, VAR VARP, TopOne String Functions Len, Concat, CharIndex Substring, Lower Upper, PatIndex Temporal Functions Lag, IsFirst,Last CollectTop Mathematical Functions ABS, CEILING, EXP, FLOOR POWER, SIGN, SQUARE, SQRT

18 Geospatial functions

19 Scenario – Twitter Analytics IDCreatedAtUserNameTimeZoneTextLanguageTopic 1 2015-04- 30T20:45:30 Joshua X Eastern Time (US & Canada) Love my @Xbox updatesenXBox 2 2015-04- 30T20:45:31 Cristabel YLondon RT @verge: Streaming Xbox One games.. enXBox … “A news media website wants to increase site traffic by covering trending topics on social media.” To determine which topics are immediately relevant to customers, they need real-time analytics about the tweet volume and sentiment for each topic. Twitter payload

20 SELECT count(*), Topic FROM InputStream GROUP BY Topic, TumblingWindow(second, 5)

21 Filters SELECT UserName, TimeZone FROM InputStream WHERE Topic = 'XBox' Show me the user name and time zone of tweets on the topic XBox "Haroon”, “Eastern Time (US & Canada)” "XO", “London” “Zach Dotseth“, “London”, “Football”,(…) "Haroon”, “Eastern Time (US & Canada)” “XBox”,(…) "XO",”London”, “XBox“, (…)

22 SELECT CreatedAt, UserName, Language= CASE Language WHEN 'en' THEN 'English' ELSE 'Non English' END FROM InputStream Tell me the time, user name and if the language is English or not "2014-11-04T23:55:10", “Zach Dotseth“, “English” “2014-11-04T23:55:11", "Haroon“, “Non English” "2014-11-04T23:55:12", "XO”, “English” "2014-11-04T23:55:10", “Zach Dotseth“, “en”, (…) “2014-11-04T23:55:11", "Haroon”, “pk", (…) "2014-11-04T23:55:12", "XO", “en“, (…)

23 Windows can be – Hopping, Sliding or Tumbling Windows are fixed length Must be used in a GROUP BY clause 1542686 4 t1t2t5t6t3t4 Time Window 1Window 2Window 3 Aggregate Function (Sum) 1814 Output Events

24 SELECT TimeZone, COUNT(*) AS Count FROM InputStream TIMESTAMP BY CreatedAt GROUP BY TimeZone, TumblingWindow(second,10) Every 10 seconds give me the count of tweets over the last 10 seconds 1542686 5 05201015 Time (seconds) 15426 86 25 A 10-second tumbling window 30 361 5361

25 SELECT Topic, COUNT(*) AS TotalTweets FROM TwitterStream TIMESTAMP BY CreatedAt GROUP BY Topic, HoppingWindow(second, 10, 5) Every 5 seconds give me the count of tweets over the last 10 seconds 1542687 05201015 Time (secs) 25 A 10-second Hopping Window with a 5-second “Hop” 30 426 86 5361 15 426 8653 61 53

26 SELECT Topic, COUNT(*) FROM TwitterStream TIMESTAMP BY CreatedAt GROUP BY Topic, SlidingWindow(second, 10) HAVING COUNT(*) > 10 Give me the count of tweets for all topics which are tweeted more than 10 times in the last 10 seconds 1 5 05201015 Time (secs) 25 A 10-second Sliding Window 51 5 1 Entry Exit 15

27 Event when store is within a possible flooding zone SELECT Store.Polygon, Flooding.Polygon FROM Cars c JOIN Flooding f ON ST_OVERLAPS(s.Polygon, f.Polygon ) Storm is heading my way SELECT Cars.Location, Storm.Course FROM Cars c JOIN Storm s ON ST_OVERLAPS(c.Location, s.Course) Geospatial examples Combination of clustering and heat maps. Clusters are represented using color coded geometric shapes that fit together evenly

28 {“XO”, 4, “Win10”}{“Jo”, 0, “Surface”} {“Foo”,4, “Bing”} {“Dip”, 2, “XBox”} {“XO”, 0, “Win10”} {“Dip”, 0, “Xbox”} {“Jo”, 4, “Surface”} {“Foo”, 0, “Bing”} Twitter Stream: (same stream, further down the timeline) SELECT TS1.UserName, TS1.Topic FROM TwitterStream TS1 TIMESTAMP BY CreatedAt JOIN TwitterStream TS2 TIMESTAMP BY CreatedAt ON TS1.UserName = TS2.UserName AND TS1.Topic = TS2.Topic AND DateDiff(second, TS1, TS2) BETWEEN 1 AND 60 WHERE TS1.SentimentScore != TS2.SentimentScore “List all users and the topics on which they switched their sentiment within a minute“

29 Show me if a topic is not tweeted for 10 seconds since it was last tweeted. SELECT TS1.CreatedAt, TS1.Topic FROM TwitterStream TS1 TIMESTAMP BY CreatedAt LEFT OUTER JOIN TwitterStream TS2 TIMESTAMP BY CreatedAt ON TS1.Topic = TS2.Topic AND DATEDIFF(second, TS1, TS2) BETWEEN 1 AND 10 WHERE TS2.Topic IS NULL {“XO”, 4, “Win10”}{“WAA”, 2, “Microsoft”} {“AB”, 0, “Bing} {“Dip”, 4, “Xbox”} {“Foo”, 0, “Win10”}{“Tim”, 2, “Microsoft”} {“AB”, 0, “Bing”} Twitter Stream:

30 Reference Data SELECT myRefData.Name, myStream.Value FROM myStream JOIN myRefData ON myStream.myKey = myRefData.myKey

31 Stream Analytics Partitioning allows for parallel execution over scaled-out resources SELECT Count(*) AS Count, Topic FROM TwitterStream PARTITION BY PartitionId GROUP BY TumblingWindow(minute, 3), Topic, PartitionId QueryResult 1 QueryResult 2 QueryResult 3 Event Hub

32

33

34 ML Function callouts In Private Preview! Azure ML can publish web endpoints for operationalized models Azure Stream Analytics can bind custom function names to such web endpoints SELECT text, sentiment(text) AS score FROM myStream

35

36

37 Input and Output Management Transformations Management Programmatic Access with REST APIs Jobs Management Start Job Stop Job Create Job Delete Job List Jobs Update Job Create Input / Output Delete Input / Output List Input / Output Update Input / Output Create Transformation Delete Transformation Get Transformation Update Transformation The full functionality of Azure Stream Analytics is through REST APIs. Enables programmatic access Useful for automation through scripting Embed in other applications/tools

38 Benefits  Greener Buildings  Comfortable occupants “The queries we need to run are quite complicated.... We are able to do this much quicker with Azure Stream Analytics, and with very low overhead.” - Arvind Shetty, Technology Specialist “The queries we need to run are quite complicated.... We are able to do this much quicker with Azure Stream Analytics, and with very low overhead.” - Arvind Shetty, Technology Specialist Making buildings smarter

39 Benefits  Improved patient care  Newer revenue models Opening new opportunities in healthcare After the transaction “The ability to collect vital telemetry data from deployed devices has been a key objective of ours. A cloud-based solution allowing us to collect data on device performance in real-time helps us to be more proactive in our customer support and ensure that our NIOX devices help the physicians deliver the best possible outcome for the patient,” - Scott Myers, CEO “The ability to collect vital telemetry data from deployed devices has been a key objective of ours. A cloud-based solution allowing us to collect data on device performance in real-time helps us to be more proactive in our customer support and ensure that our NIOX devices help the physicians deliver the best possible outcome for the patient,” - Scott Myers, CEO

40 Benefits  Run successful promotions in real time  Better customer service Helping retailers increase sales through real time analytics “As NEC continues to expand its strong solution portfolio to customers globally.. the power of NEC's biometrics technologies combined with Azure based cloud-based services can be delivered to our customers across geographic boundaries” - Naohiko Torii, Senior Manager “As NEC continues to expand its strong solution portfolio to customers globally.. the power of NEC's biometrics technologies combined with Azure based cloud-based services can be delivered to our customers across geographic boundaries” - Naohiko Torii, Senior Manager

41 Benefits  More healthy choices for patients  Newer revenue opportunities Innovative farming helps improve patient lives “By leveraging the Microsoft Azure IoT and the Fujitsu platform, we are able to deliver real-time visualization of big data analytics to improve the entire production” - Hiroyuki Sakai, Vice President “By leveraging the Microsoft Azure IoT and the Fujitsu platform, we are able to deliver real-time visualization of big data analytics to improve the entire production” - Hiroyuki Sakai, Vice President

42 Devices Microsoft IoT suite Back end systems and processes Event Hub Storage blobs Document DB Web/Mobile App Stream Analytics Logic Apps Azure Active Directory IoT Hub Web Jobs

43 Microsoft Cortana Intelligence Suite Action People Automated Systems Apps Web Mobile Bots Intelligence Dashboards & Visualizations Cortana Bot Framework Cognitive Services Power BI Information Management Event Hubs Data Catalog Data Factory Machine Learning and Analytics HDInsight (Hadoop and Spark) Stream Analytics Intelligence Data Lake Analytics Machine Learning Big Data Stores SQL Data Warehouse Data Lake Store Data Source s Apps Sensors and devices Data


Download ppt "Real-time Fraud DetectionStreaming ETLPredictive MaintenanceCall Center Analytics IT Infrastructure and Network Monitoring Customer behavior."

Similar presentations


Ads by Google