C* Summit 2013: Big Data Analytics – Realize the Investment from Your Big Data Clusters by Mark Davis
2 likes1,648 views
The document discusses the role of big data analytics in various industries and its impact on decision-making and operational efficiency. It highlights specific use cases such as military intelligence, competitive analysis, drug discovery, customer relationship management, fraud detection, and consumer behavior analysis. Additionally, it outlines Dell's Kitenga Analytics Suite aimed at transforming big data into actionable intelligence through various analytical and visualization techniques.
C* Summit 2013: Big Data Analytics – Realize the Investment from Your Big Data Clusters by Mark Davis
- 1. Big Data Analytics – Realize the Investment from Your Big Data Clusters Mark Davis| Senior Architect and Principal Engineer, Dell Inc.
- 2. Big Data and Society How Is Big Data Affecting Our World?
- 3. 200EB = 1018 B 1ZB = 1021 B 10EB 100TB 2000198519001750 Industrial Revolution #1 Industrial Revolution #2 Industrial Revolution #3 Industrial Revolution #4 R. J. Gordon: Is US economic growth over? Faltering innovation confronts the six headwinds. CEPR Policy Insight No 63
- 4. Distributed File System MapReduce Eventually Consistent Column Store Analytics Database NoSQL Structured Semi-structured Unstructured Text Analytics Machine Learning The Big Data “Zoo”
- 5. Big Data Use Cases How Is Big Data Being Consumed Today?
- 6. SourcesKAS GOAL: Improve force effectiveness SOURCES: Situation reports and acquired multi- source intelligence ANALYSIS: Extract named entities and relationships, classify and label, normalize geospatial and temporal metadata; visually understand relationships and trends ACTION: Identify mission objectives and create priorities Defense Intelligence Visualization metadata relationships data Visual Understanding entities
- 7. * Current system doesn’t scale * Oracle with text plug-in * Overwhelmed by intelligence needs * Need analytic capability with search US Army
- 8. SourcesKAS GOAL: Be more competitive SOURCES: Patents, PR announcements, legal documents, whitepapers, crawled websites ANALYSIS: Extract named entities and relationships, classify and label; visually understand relationships and trends ACTION: Change R&D priorities and improve marketing approaches Competitive Intelligence Viz/Search metadata relationships data Understanding entities
- 9. * Understand IP among competitors * Assist legal team with litigation * Custom search experience * Custom extractors: Electronic parts Memory types Flash memory Customer: Technology Company
- 10. SourcesKAS GOAL: Discover new drugs, detect side-effects, speed R&D SOURCES: Published research reports, patents, adverse effects databases, genomics and proteomics databases ANALYSIS: Extract named entities and relationships, classify and label; visually discover trends and relationships ACTION: Change R&D priorities Drug Discovery Viz/Seach relationships data Understanding entities pathways sequences
- 11. * Lousy search * Internal regulators can’t find by accession number * Custom extractors: Accession number Ontology of active ingredients Drug names FDA
- 12. SourcesKAS GOAL: Scalable analysis of customer relationship engagements SOURCES: Call center and web help contact narratives ANALYSIS: Ingest massive data sets; visually discover trends, novelty, and relationships ACTION: Predict new product issues CRM Analytics Viz/Search relationships data Understanding My iPhone is very hot…
- 13. SourcesKAS GOAL: Scalable analysis of network failures SOURCES: Uploaded syslog data and configuration for routers and switches ANALYSIS: Ingest massive data sets; visually discover trends and relationships ACTION: Solve network problems Network Analytics Viz/Search relationships data Understanding
- 14. * Unable to manage customer network signals * RDBMS * Tiger team dumps database and runs Perl scripts for analysis Router/Switch Vendor
- 15. SourcesKAS GOAL: Reduce fraud SOURCES: Analysis customer data ANALYSIS: Extract patterns of web and service usage, classify, label with normalized geospatial and temporal metadata; visually understand relationships and trends. ACTION: Indentify fraudulent transactions and patterns Financial Services: Fraud Viz/Search metadata relationships data Understanding
- 16. SourcesKAS GOAL: Identify what people want to buy SOURCES: Crawl Twitter, blogs, and websites ANALYSIS: Extract sentiments about products, classify, label with normalized geospatial and temporal metadata; visually understand relationships and trends. ACTION: Target sales and enhance offerings Buy Signals Viz/Search metadata relationships data Understanding sentiments
- 17. SourcesKAS GOAL: Find case-supporting and actionable information SOURCES: Email repositories, Office documents, patents, memos ANALYSIS: Extract named entities and relationships, classify and label; visually discover trends and relationships ACTION: Develop legal theories and prepare for arguments Legal Informatics Viz/Search metadata relationships data Understanding entities
- 18. Dell’s Kitenga Analytics Suite
- 19. Aggregate Count Extract Transform Chart Graph Model Visualize Search Predict Transform Big Data into Actionable Intelligence
- 20. Search Facetted Search, Visualization Analytics Extract, Crawl, Index, NLP, Transform, Machine Learning Analytical Producer Analytical Consumer Visualization Visualize, Model, Interact
- 27. Cassandra in the Zoo How Dell Is Integrating Cassandra
- 29. THANK YOU