Download to read offline
DATA BASE MANAGEMENT IN CLOUD - UNIT -1 PPT
- 1. DATA MANAGEMENT IN CLOUD Dr.Abdur Rahman, AP/CSE
- 2.
- 3.
- 4. Distributed Database • Database- Organized collection, structured data. • Distributed – Shared, Spread out. • Collection of data that is shared which is physically distributed over a computer network on different sites. • Logically it’s a single DB divided into number of pieces called Fragments. • Distributed databases use a client/server architecture to process information requests.
- 5. Types of DistributedDatabase • Homogenous Distributed Database Systems • Heterogeneous Distributed Database Systems • Generic connectivity • ODBC • OLE DB Protocols
- 6. Concepts • Database Links •Defines a one-way communication path from server to another database server. Types of DB Links • Public • Private • Connected
- 7. Distributed Data Storage •Replication • Redundantly at 2 or ore sites. • Increased Availability • Parallel Processing • Constant Updates - Overhead • Fragmentation • Divided into smaller parts. • Consistent • No redundancy
- 8. Distributed Databases Features •Location independent • Distributed query processing • Distributed transaction management • Hardware independent • Operating system independent • Network independent • Transaction transparency • DBMS independent
- 9. Examples of distributeddatabases • Apache Ignite, • Apache Cassandra, • Apache HBase, • Couchbase Server, • Amazon SimpleDB, • Clusterpoint, and • FoundationDB.
- 10. Data Allocation • Intelligentdistribution of your data pieces. • Performance and Availability • Types: Centralized, Partition, and Replicated. • Strategies: • Data Fragmentation • Dividing the database into part/sub-table. • Horizontal fragmentation. • Vertical fragmentation, • Mixed or Hybrid fragmentation. • Data Replication • Copying of Data – Multiple locations.
- 11. Horizontal Fragmentation ID NameAge Marks 1 A 21 20 2 B 22 25 3 C 23 30 4 D 24 35 SELECT * FROM student WHERE marks < 35; ID Name Age Marks 1 A 21 20 2 B 22 25 SELECT * FROM student WHERE marks > 35; ID Name Age Marks 4 D 24 35 T1 T2 Types: •Primary •Derived •Complete T = T1 ∪ T2 ∪ …. ∪ TN
- 12. Vertical Fragmentation SELECT NameFROM Table; ID Name Age Marks 1 A 21 20 2 B 22 25 3 C 23 30 4 D 24 35 SELECT Age FROM Table; Name A B C D Age 21 22 23 24
- 13. Hybrid Fragmentation SELECT NameFROM Table WHERE age = 22; ID Name Age Marks 1 A 21 20 2 B 22 25 3 C 23 30 4 D 24 35 Name B
- 14. Types of DataReplication • Transactional Replication • A complete copy of your database • Copies of new data changes • Database are synced in real-time • Snapshot Replication • Simplest type of Data Replication • Current state at a specific in time • Merge Replication • Tracks subsequent data changes and schema modifications • Synchronizes using merge agents
- 15. Structure of DistributedDatabase
- 16. Architecture Models ofDistributed Database Systems • Client-Server Architecture • Peer-to-peer Architecture • Multi DBMS Architecture
- 17. Trade-off in DistributedDatabase • Trade-off • CAP Theorem • Consistency: Every read receives the most recent write or an error. • Availability: Every request receives a response • Partition tolerance: Continued Functioning.
- 18. Practical Implications • Understandthe requirements • Choose the appropriate replication scheme • Use appropriate data structures • Plan for failures “The fate of your distributed system rests on your ability to make the right trade-offs. Choose wisely! “
- 19. Objectives of theDesign of Data Distribution • Processing locality- placing data as close as possible to the applications which use them. • Availability and reliability of distributed data. • Workload distribution. • Storage costs and availability.
- 20. Distributed Database Design- Concept • Centralized DB Issue • Designing the "conceptual schema" - High Level Description - Main Concept and Relationships. • Designing the "physical database," i.e., mapping the conceptual schema to storage areas and determining appropriate access methods. • Distributed DB Issue • Designing the fragmentation. • Designing the allocation of fragments- mapping to physical image.
- 21. Two Strategies • TopDown • Designing systems from scratch • Mostly in homogeneous systems
- 22. Bottom Up Approach •When the databases already exist at a number of sites
- 25. Design of DistributedDatabase • Top Down Design • Designing the global schema, and we • Designing the fragmentation of the database, and then by • Allocating the fragments to the sites, and • Creating the physical images • Bottom Up • The selection of a common database model for describing the global schema of the database. • 2. The translation of each local schema into the common data model. • 3. The integration of the local schemata into a common global schema.Loosely Coupled System