1. Trang chủ
  2. » Công Nghệ Thông Tin

Distributed Database Management Systems: Lecture 13

18 2 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 18
Dung lượng 169,72 KB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

Distributed Database Management Systems: Lecture 13. The main topics covered in this chapter include: interests of the local and head offices mar result following vertical partitions of this table: EMP1(eId, eName, eDept); EMP2(eId, eQual, eSal); degree of fragmentation; correctness rules for fragmentation;...

Trang 1

Distributed Database Management Systems

Lecture 13

Trang 2

Virtual University of Pakistan 2

How

Interests of the local and head offices mar result following vertical

partitions of this table:

EMP1(eId, eName, eDept) EMP2(eId, eQual, eSal)

Trang 3

Virtual University of Pakistan 3

1- Vertical; Different subsets of

attributes are stored at different

places, like,

Table EMP(eId, eName, eDept, eQual, eSal)

Trang 4

Virtual University of Pakistan 4

Interests of the local and head offices may result following vertical partitions of this

table:

EMP1(eId, eName, eDept)

EMP2(eId, eQual, eSal)

Trang 5

Virtual University of Pakistan 5

CLIENT

(cAC#, cName, cAdr, cBal) table is

placed in different databases based on their location, like from Lahore, Pindi, Karachi, Peshawar, Quetta

Trang 6

Virtual University of Pakistan 6

3- Degree of Fragmentation

Between no to the extreme level that could be to the individual tuple or

column level; a compromised decision

Trang 7

Virtual University of Pakistan 7

4- Correctness Rules for

Fragmentation

Trang 8

Virtual University of Pakistan 8

• Completeness: each of the data item

(a tuple or a attribute) that can be in R can also be in one or more Ri

∀ x R, Rx R, R∈∈ ∃∃ i such that x R such that x R∈∈ i

Resembles Lossless decomposition

property in Normalization

Trang 9

Virtual University of Pakistan 9

• Reconstruction: it should be

possible to define a relational operator such that the original relation can be reconstructed

R = g(R1, R2, …, Rn)

Reconstruction ensures that the dependencies of the original

relation are preserved in

fragmentation

Trang 10

Virtual University of Pakistan 10

• Disjoint ness:

if data item x is in Rj, it is not in any other fragment

∀ x Rx R∈∈ i, R R∃∃ j such that x R such that x R∈∈ j, i ≠ j

Trang 11

Virtual University of Pakistan 11

• Allocation Strategy:

Partitioned, fully or partially replicated; depends mainly on requirements

Trang 12

Virtual University of Pakistan 12

Horizontal Fragmentation

Trang 13

Virtual University of Pakistan 13

• is performed based on some

Predicate/ Condition

Trang 14

Virtual University of Pakistan 14

Primary Horizontal Fragmentation(PHF)

Trang 15

Virtual University of Pakistan 15

Information Requirements

Trang 16

Virtual University of Pakistan 16

Database Information: We may

need to consult the conceptual

DB design

RDM being Semantically not that rich

Apart from tables, we need

relationships, cardinality and the owner and member tables

Trang 17

Virtual University of Pakistan 17

owner = PAY member = EMP title, sal

eNo, jNo, resp, dur

PAY

EMP

ASIGN

PROJ

Trang 18

Virtual University of Pakistan 18

Application Requirement

Ngày đăng: 05/07/2022, 13:33

TÀI LIỆU CÙNG NGƯỜI DÙNG

  • Đang cập nhật ...

TÀI LIỆU LIÊN QUAN