AcidBaseConundrum.pdf

When we start a new project, we are often posed with the task of setting up a data store. The arises the inevitable question "Should I use a SQL based store or NOSQL one". We might tend to lean towards SQL due to it's familiarity and ubiquity; or lean towards NoSQL due to its ease and low setup cost. In this talk, we will look at database choices through a different lens. Taking inspiration from CAP theorem we will understand the ACID/BASE properties of datatbases that will act as guiding principles for choosing database.

Transcript

1. The ACID/BASE Conundrum Sangeetha KP (sangee_kp)
2. Cassandra or Spanner
3. SQL vs NoSQL Cassandra or Spanner
4. Cassandra or Spanner SQL vs NoSQL ACID vs BASE
5. The ACID/BASE Conundrum Sangeetha KP (sangee_kp)
6. Single Node System Multi Node System (Distributed System)
7. CONSISTENCY
8. CONSISTENCY AVAILABILITY
9. CONSISTENCY AVAILABILITY PARTITION TOLERANCE
10. CAP Theorem It is impossible for a to simultaneously provide guarantees: 1. Consistency 2. Availability 3. Partition Tolerance distributed data store more than two out of the following three C CA P A CP AP
11. Atomicity Consistency Isolation Durability Banking Transactions
12. Atomicity Consistency Isolation Durability Basically Available Soft state Eventual Consistency Banking Transactions Twitter Feed
13. x=5 x=10 x=20 x=5 ACID Compliance
14. x=5 x=20 ACID Compliance x=10
15. x=10 x=10 x=20 ACID Compliance
16. x=5 x=5 x=20 BASE Compliance x=10
17. x=5 x=5 x=5 x=20 BASE Compliance x=10 x=5
18. x=10 x=20 x=10 BASE Compliance x=20
19. x=20 x=20 x=20 BASE Compliance x=20
20. SQL NoSQL NewSQL Google Spanner, CockroachDB, VoltDB ACID + Scalability
21. NewSQL is a class of relational database management systems that seek to provide the scalability of NoSQL systems for online transaction processing (OLTP) workloads while maintaining the ACID guarantees of a traditional database system.
22. Reference materials: https://tinyurl.com/acidbase Illustration: https://undraw.co/ sangee_kp Thank you!