Course Features
- Lectures 24
- Quizzes 0
- Duration 3 zile
- Skill level All levels
- Language Romana
- Students 12
- Assessments Yes
-
Day 1: Big Data architecture review and intro of potential storage components (file and database storage)
Role of distributed storage layer in a big data architecture
- Components of the storage layer: Distributed file storage (HDFS & other file storage), distributed databases
- noSQL Overview: Distributed databases (common concepts); Consistency & the CAP Theorem; noSQL solutions overview
- Distributed File storage: HDFS overview, HDFS and Computing and Analysis layer of Big Data architectures, Hadoop architecture components
- Analysis layer overview
- Apache Hive, Impala, Spark SQL overview
- HDFS & Hive overview (Create tables (internal/external), SERDE’s, partitions, buckets, file formats, …)
-
Day 2: Columns store noSQL: Apache Cassandra
- The Column-Family Data Model: why better for specific use cases
- Main concepts: data partitioning, distribution, replication and consistency, write and read paths (cluster and node level)
- Architecture: gossip, snitch, multi DC deployments
- Data model: keyspace, table, partition key(s), clustering key(s), indexes
- Cassandra Query Language (create keyspace, table, insert/update data into tables, delete, query data)
-
Day 3: Columns store noSQL: Apache HBase
- Hbase most known use cases, differences vs Apache Cassandra
- Architecture: master, meta, zookeeper, region servers, HDFS
- Basic concepts: row-key, tables, regions, persistence of data, write and read paths
- Regions scaling and auto-scaling
- Data Model
- Row Key Design
- Operations: write, read and delete data
- Documents store noSQL: MongoDB
- The Document Data Model (documents, collections)
- MongoDB Use Cases
- Architecture: Replication via Replica Sets; MongoDB and the CAP Theorem; Auto-Sharding, Shard Keys, and Horizontal Scalability
- DML
- Indexing
