Distributed Database Systems-Data Localization.pptx
Download as PPTX, PDF0 likes124 views
The document discusses data localization in distributed database systems, focusing on various fragmentation techniques such as vertical, derived horizontal, and hybrid fragmentation. It explains how these methods optimize data distribution and improve query processing, detailing examples for better understanding. The goal is to effectively support operations involving projection, selection, and join, balancing efficiency among different query types.
Distributed Database Systems-Data Localization.pptx
- 2. Lecture Topics Data Localization Distributed Database Systems Lecture #20 By Muhammad Iqbal Superior College Mian Channu Computer Graphics
- 3. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems
- 4. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 2. Reduction for Vertical Fragmentation
- 5. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 1 Reduction for Vertical Fragmentation The vertical fragmentation function distributes a relation based on projection attributes. Since the reconstruction operator for vertical fragmentation is the join, the localization program for a vertically fragmented relation consists of the join of the fragments on the common attribute. For vertical fragmentation, we use the following example.
- 6. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 1 Reduction for Vertical Fragmentation Example 7.11. Relation EMP can be divided into two vertical fragments where the key attribute ENO is duplicated:
- 7. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 1 Reduction for Vertical Fragmentation
- 8. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 3 Reduction for Derived Fragmentation • Derived horizontal fragmentation is another way of distributing two relations so that the joint processing of select and join is improved. • Typically, if relation R is subject to derived horizontal fragmentation due to relation S, the fragments of R and S that have the same join attribute values are located at the same site. • In addition, S can be fragmented according to a selection predicate.
- 9. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 3 Reduction for Derived Fragmentation The localization program for a horizontally fragmented relation is the union of the fragments. In our example, we have ASG = ASG1∪ ASG2
- 10. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 4 Reduction for Hybrid Fragmentation • Hybrid fragmentation is obtained by combining the fragmentation functions discussed above. • The goal of hybrid fragmentation is to support, efficiently, queries involving projection, selection, and join. • Note that the optimization of an operation or of a combination of operations is always done at the expense of other operations. • For example, hybrid fragmentation based on selection-projection will make selection only, or projection only, less efficient than with horizontal fragmentation (or vertical fragmentation). • The localization program for a hybrid fragmented relation uses unions and joins of fragments.
- 11. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 4 Reduction for Hybrid Fragmentation In our example, the localization program is EMP = (EMP1 ∪ EMP2) ENO EMP3 ✶
- 12. Data Localization by Muhammad Iqbal Superior College Mian Channu Distributed Database Systems 4 Reduction for Hybrid Fragmentation SELECT ENAME FROM EMP WHERE ENO="E5"
- 13. Allah Hafiz