The Changing World of Business Intelligence
• Big Data: Hype or reality?
• Operational intelligence: does it require online data warehouses?
• Fast data is the next frontier of big data
• Data warehouses in the cloud
• Self-service BI
• The business value of analytics
• The relationship between big data and analytics
• The Hadoop software stack explained, including HDFS, MapReduce, YARN, Hive, Storm, Sqoop, Flume, and HBase
• The balancing act: productivity versus scalability
• Making big data available to a larger audience with SQL-on-Hadoop engines, such as Apache Drill, Apache Hive, Apache Phoenix, Cloudera Impala, HP Vertica, IBM BigSQL, JethroData, MemSQL, SparkSQL, and Splice Machine
• Spark is about in-memory analytical processing
• The interfaces: SQL, R, Scala, Python
• Does Spark need Hadoop?
• The relationship between Spark and data science
• Examples of use cases of Spark
• Classification of NoSQL database servers: key-value stores, document stores, column-family stores and graph data stores
• Market overview: CouchDB, Cassandra, Cloudera, MongoDB, and Neo4j
• Strong consistency or eventual consistency?
• Why an aggregate data model?
• Use case of NoSQL products
• How to analyze data stored in NoSQL databases
Overview of Analytical SQL Database Servers
• Are classic SQL database servers more suitable for data warehousing?
• Important performance improving features: column-oriented storage, in-database analytics
• Market overview of analytical SQL database servers: Apache Greenplum, Exasol, HP Vertica, IBM PureData Systems for Analytics, InfoBright, JustOneDB, Kognitio WX2, Microsoft PDW, Oracle In-Memory, SAP HANA and Sybase IQ, SnowflakeDB, Teradata Appliance, and Teradata Aster Database
Technologies for Fast Data and Streaming Analytics
• The key use case for fast data: the Internet of Things (IoT)
• IoT implies streaming data and fast analysis of data – analytics at the speed of business
• IoT devices: Smartphones (watches), RFID sensors, machines, general sensors, cameras, pace makers, and so on
• The challenge: real-time reactions on streaming data
• The difference between big data and fast big data
• Technologies for streaming data: Apache Kafka, Apache ActiveMQ, Amazon Kinesis, Kestrel, RabbitMQ, and ZeroMQ
• Differences between these new technologies and traditional message queuing products
• Products for big data streaming: Apache Storm and Flink, IBM InfoSphere Streams, Informatica for Streaming Analytics, Software AG Apama, and Spark Streaming
• How to integrate fast data with the enterprise data warehouse?
Data Virtualization for Agile BI systems and Lean Integration
• Data virtualization offers on-demand data integration
• Seamlessly integrating big data and the data warehouse
• Market overview: AtScale, Cirro Data Hub, Cisco Information Server, Denodo Platform, RedHat JBoss Data Virtualization, Rocket DV, and Stone Bond Enterprise Enabler
• Importing non-relational data, such as XML documents, web services, NoSQL and Hadoop data, and unstructured data
• Differences between data virtualization and data blending
New Business Intelligence Architectures
• Discussion of different BI architectures, including Kimball’s Data Warehouse Bus, Architecture, Inmon’s Corporate Information Factory, DW 2.0, the Federated Architecture, the Centralized Warehouse Architecture, the Data Virtualization Architecture, and the BI in the Cloud Architecture
• Do we still need data marts?
• What is the role of master data management in BI architectures?
• Using data vault to create more flexible data warehouses
• Data warehouse automation to create data warehouses and data marts faster
NewSQL Database Servers
• NewSQL stands for high-performance transactional SQL database servers
• Simpler transaction mechanisms to implement scale-out
• What does the term geo-compliancy mean?
• Market overview: Clustrix, GenieDB, NuoDB, and VoltDB
Data Modelling for Big Data, Hadoop, and NoSQL
• Explanation of non-relational concepts, such as column families, hierarchies, sets, and lists
• Is storing unstructured and semi-structured data really more flexible?
• The differences between schema-on-read and schema-on-write
• Rules for transforming classic data models to NoSQL concepts
• Application needs influence database design