This session will talk about the need and evolution of NoSQL, fundamentals and comparison of NoSQL and RDBMS systems. We will try to understand the practical relevance of NoSQL with a real problem statement and it’s solution based on NoSQL. The session will briefly also look at various implementations and types of NoSQL solutions.
Session Outline:
- What is NoSQL
- Need of NoSQL:
- Non Relational in Nature : In a NoSQL database, there is no fixed schema and no joins. Schema Managment on systems where downtime is a complete no-no is a difficult task with relational databases. So systems which have reduced schema compelxity helps catering this limitation.
- Distributed/Scalability : Horizontoal Scalability Vs Vertical . Handling big data
- High Availability : Refers to availability and reliability of the system. This demands 100% operational or never failimg systems. No SPOF
- Eventually Consistent
- Cost
- Some basic Fundamentals/Theories which help us in designing or implementing large scale databases and helps understanding the constraint cum limitation’s.
- CAP theorem
-
BASE Versus ACID
-
Consistent Hashing (HashRing)
-
Types of NoSQL Databases
-
Wide Column Store / Column Families : examples: Cassandra/Hadoop(Hbase)/
Hypertable/Cloudera etc. -
Document Based Store’s : MongoDb/CouchDb/TerraStore etc.
-
Key/Value(Tuple Store) Store : Redis,MemcacheDb/Voldemort etc.
-
Graph Data Store : Neo4J/GraphBase/FlockDb etc
-
-
Use Case for NoSQL:The session would cover a practical use case and how it get’s solved with a NoSQL solution cassandra. Stuff in terms of how to model a column family as per the use case, schemaless advantage , client etc .It will also talk of operational/infrastructure hiccups in terms of configuring multi data center ring , adding removing nodes at runtime , core’s and memory requirement from cassandra etc to be done in practice.
-
Limitations of NoSQL Solutions
-
Maturity
-
Support
-
Skillset and Administration/Operation
-
Analytics and BI support
-









