SlideShare ist ein Scribd-Unternehmen logo
1 von 22
Downloaden Sie, um offline zu lesen
Concepts of Parallel and Distributed Database Systems
Key concepts:
- concept and structure of parallel and distributed databases
- motivations for parallelism and distribution
- types of parallelism
- benefits of DDBMS
- drawbacks of DDBMS
- Date’s rules
- local autonomy
- function distribution
- continues operation
- location independence
- fragmentation independence
- replication independence
- overview of query processing
- overview of distributed transaction management
- hardware independence
- DBMS independence
- Distribution transparency
- Replication transparency
- Fragmentation transparency
- history of parallel databases
- history of distributed databases
Parallel and Distributed Databases
•A parallel database aims principally linear speedup and
scaleup:
Linear scaleup refers to a sustained performance for a
linear increase both in database size and processing and
storage power.
Linear speedup refers to a linear increase in
performance for a constant database size.
•Sharing data is the key of a distributed DB →
cooperative approach
Since data is distributed, users that share that data can
have it placed at the site they work on, with local
control (Local autonomy)
•Distributed and parallel databases improve reliability
and availability
I.e. they have replicated components and thus eliminate
single points of failure.
•The distribution of data and the parallel/distributed
processing is not visible to the users → transparency
•Distributed Database (DDB)
A collection of multiple, logically interrelated
databases distributed over a computer network
•Distributed Database Management System
(D-DBMS)
The software system that permits the management of
the DDB and provides an access mechanism that makes
this distribution transparent to the users.
•Distributed Database System (DDBS)
DDBS = DDB + D-DBMS
•What is not a DDBS: A database system which resides
on
•a timesharing computer system
•a loosely or tightly coupled multiprocessor system
•one of the nodes of a network of computers - this is a
centralised database on a network node
Reasons of data distribution
•Improved reliability and availability through distributed
transactions
•Improved performance
•Allowing data sharing while maintaining some measure
of local control
•Easier and more economical system expansion
•Distributed nature of some database applications
Additional functionality of a D-DBMS
•Distribution leads to increased complexity in the system
design and implementation
•D-DBMS must be able to provide some additional
functionality to those of a central DBMS:
• To access remote sites and transmit queries and data among the
various sites via a communication network.
• To keep track of the data distribution and replication in the
DDBMS catalog.
• To devise execution strategies for queries and transactions that
access data from more than one site.
• To decide on which copy of a replicated data item to access.
• To maintain the consistency of copies of a replicated data item.
• To maintain the global conceptual schema of the distributed
database.
• To recover from individual site crashes and from new types of
failures such as failure of a communication link.
Distributed DBMS Issues
•Distributed Database Design
•how to distribute the database
•replicated & non-replicated database distribution
•a related problem in directory management
• Query Processing
•convert user transactions to data manipulation instructions
•optimisation problem
•min{cost = data transmission + local processing}
•general formulation is NP-hard
•Concurrency Control
•synchronisation of concurrent accesses
•consistency and isolation of transactions' effects
•deadlock management
•Reliability
•how to make the system resilient to failures
•atomicity and durability
Date’s 12 rules for distributed database systems
Rule 0: to the user, a distributed system should look
exactly like a non-distributed system
other rules:
1. Local autonomy
2. No reliance on central site
3. Continuous operation
4. Location independence
5. Fragmentation independence
6. Replication independence
7. Distributed query processing
8. Distributed transaction management
9. Hardware independence
10. Operating system independence
11. Network independence
12. DBMS independence
The great principles are:
Autonomy, Independence and Transparency.
Rule 1: Local Autonomy
Autonomy objective: Sites should be autonomous to the
maximum extent possible.
•Local data is locally owned and managed, with local
accountability
•security considerations
•integrity considerations
•Local operations remain purely local
•All operations at a given site are controlled by that site;
no site X should depend on some other site Y for its
successful functioning
Otherwise, if site Y is down, site X could not carry out
the operation although there is nothing wrong at site X.
•In some situations some slight loss of autonomy is
inevitable
•fragmentation problem – rule 5
•replication problem – rule 6
•update of replicated relation – rule 6
•multiple-site integrity constraint problem – rule 7
•a problem of participation in a two-phase commit process –
rule 8
In consequence: All data ‘really’ belongs to some local
database even it is accessible from remote.
Rule 2: No reliance on central site
There must not be any reliance on a central ‘master’site
for some central service, such as centralised query
processing or centralised transaction management, such
that the entire system is dependent on that central site.
•Reliance on a central site would be undesirable for at
least the following two reasons:
•that central site might be a bottleneck
•the system would be vulnerable
•In a distributed system therefore, the following
functions (among others) must all be distributed:
•Dictionary management
•Query processing
•Concurrency control
•Recovery control
Rule 3: Continuous operation
There should ideally never be any need for a planned
entire system shutdown .
•Incorporating a new site into an existing distributed
system should not bring the entire system to a halt
•Incorporating a new site into an existing distributed
system should not require any changes to existing user
programs or terminal activities
•Removing an existing site from the distributed systems
should not cause any unnecessary interruptions in
service
•Within the distributed system, it should be possible to
create and destroy fragments and replicas of fragments
dynamically
•It should be possible to upgrade the DBMS at any given
component site to a newer release without taking the
entire system down
•
Rule 4: Location independence (Transparency)
Users should not have to know where data is physically
stored but rather should be able to behave – at least from a
logical standpoint – as if the data was all stored at their
own local site.
•Sometimes referenced as Transparency
•Simplifies user programs and terminal activities
•Allows data to migrate from site to site
•It is easier to provide location independence for simple
retrieval operations than it is for update operations
•Distributed data naming scheme and corresponding
support from the dictionary subsystem
•User naming scheme
•An user has to have a valid logon ID at each of multiple
sites to operate
•User profile for each valid logon ID in the dictionary
•Granting of access privileges at each component site
•
Rule 5: Fragmentation independence (Transparency)
•A distributed system supports data fragmentation if a
given relation can be divided up into pieces or
‘fragments’for physical storage purposes
•A system that supports data fragmentation should also
support fragmentation independence (also known as
fragmentation transparency)
Users should be able to behave (at least from a
logical standpoint) as if the data were in fact not
fragmented at all.
•Fragmentation is desirable for performance reasons
•Fragmentation must be defined within the context of a
distributed database
•Fragmentation independence (like location
independence) is desirable because it simplifies user
programs and terminal activities
•Fragmentation independence implies that users should
normally be presented with a view of the data in which
the fragments are logically combined together by
means of suitable joins and unions
Fragmentation principles
•Two basic kinds of fragmentation
•Horizontal fragmentation
• refers to cut between tuples
• achieved by selection
• same as data partitioning in parallel databases
• efficient for parallel scanning of the relation
•Vertical fragmentation
• refers to the cut of the schema
• achieved by a projection
• efficient if there is frequent access to different attribute groups
•The union of the horizontal fragments should be the
original relation
•The join of the vertical fragments should be the original
relation
!!! Supplemental measure must be undertaken to
guarantee that the join is nonloss!
E.g. the fragmentation of the table
Employee( Emp#, Dept#, Salary) into the fragments
Emp#, Dept# and Salary would not be valid, since there
is no common join-attribute.
Solutions are either to duplicate the key or to introduce
a hidden ‘tupleID’ where the TID is the physical or
logical address for that tuple. This TID is then included
in all fragments.
Rule 6: Replication independence (Transparency)
User should be able to behave as if the data were in fact
not replicated at all.
•A distributed system supports data replication if a given
relation (more generally, a given fragments of a
relation) can be represented at the physical level by
many distinct stored copies or replicas, at many distinct
sites.
•Replication, like fragmentation, should be “transparent
to the user”
•Replication is desirable for at least two reasons:
•Performance: applications can operate on local copies
instead of having communicate with remote sites (or a
‘nearest’copy can be fetched.
•Availability: a given replicated object remains available for
processing as long as at least one copy remains available.
•Update propagation problem!
•Fragmentation independence (like all transparency
rules) is desirable because it simplifies user programs
and terminal activities
Update propagation problem
•The major disadvantage of replication is, of course,
when a given replicated object is updated, all copies of
that object must be updated.
→ Update propagation problem.
•A common scheme for dealing with this problem is the
so-called primary copy scheme:
• One copy of each replicated object is designated as the primary
copy. The remainder are all secondary copies.
• Primary copies of different objects are at different sites (so the
distributed scheme once again).
• Update operations are deemed logically complete as soon as the
primary copy has been update. The site holding the copy is then
responsible for propagating the update to the secondary copies at
some subsequent time.
•The 'subsequent' time must be prior to COMMIT,
however, if the ACID properties of the transaction are
to be preserved →
Inconsistency ?! (see below: transaction processing).
Rule 7: Distributed query processing
It is crucially important for distributed database systems to
choose a good strategy for distributed query processing.
•Query processing in a distributed system involve
•local CPU and I/O activity at several distinct sites
•some amount of data communication among those sites
•Amount of data communication is a major performance
factor. Depending on the bandwidth and the relation
sizes the minimal communication cost strategy will be
chosen.
•Distributed query processing and query optimisation
must be aware of the distributed context:
•Where is a relation (fragment) stored?
•Are there any replicas?
•Query compilation ahead of time
•Views that span multiple sites
•Integrity constraints within a DDBS that span multiple
sites
Rule 8: Distributed transaction management
Two major aspects of transaction management, recovery
control and concurrency control, require extended
treatment in the distributed environment
•In a distributed system, a single transaction can involve
the execution of code at multiple sites and can thus
involve updates at multiple sites
•Each transaction is therefore said to consist of multiple
“agents”, where an agent is the process performed on
behalf of a given transaction at a given site
•Recovery Control: In order to ensure that a given
transaction is atomic in the distributed environment,
therefore, the system must ensure that the set of agents
for that transaction either all commit in unison or all
roll back in unison.
That effect can be achieved by means of the two-phase
commit protocol.
•Concurrency Control: is based in most distributed
systems on locking (some implemented multi-version
control ..., read can have the last version)
Recovery control: Two-phase commit protocol
•In general, the two-phase commit protocol is driven,
whenever a transaction interact with several
independent 'resource manager', i.e. each managing its
own set of recoverable resources and maintaining its
own recovery log.
•FIRST PHASE: Each site must force-write all log
entries for local resources used by the transaction out to
its physical log. Assuming the force-write is successful,
each site gives the OK.
•SECOND PHASE: If the site where the transaction has
been submitted receives all the OK form the other sites,
it force-writes an entry to its own physical log.
If anything has gone noOK, then general ROLLBACK
otherwise the decision is COMMIT DONE.
•Remark 1: If site Y acts as a participant in a two-phase
commit coordinated by site X, then Y must do what is
told by site X (roll-back) – a (minor) loss of local
autonomy.
•Remark 2: If one of the messages get lost (e.g.
ROLLBACK) then the system is an inconsistent state.
In general there can be no protocol that can guarantee
that all agents will commit unison.
Concurrency control: Primary Copy approach
•Lock means: Requests to test, set and release locks.
Update means update and acknowledgement.
•Consider for example a transaction that needs to update
an object for which exists replicas at n sites. A
straightforward implementation will require:
n lock requests; n lock grants; n update messages;
n acknowledgements; n unlock requests
makes 5*n messages (could of course be piggy-picked)
•Use the primary copy: Lock only the primary copy (one
lock request, one lock grant and one unlock) and the
site with the primary copy handles the updating (2*n
messages).
Thus we got only 2*n+3 messages.
•Remark 1: Loss of autonomy: A transaction can fail
now if a primary copy is unavailable, and a secondary
copy exists at the local site.
•Remark 2: Inconsistency is introduced if all secondary
copies are not updated prior to the COMMIT point.
Unfortunately, many commercial products support a less
ambitious form of replication in which the update propagation
is guaranteed to be done at some future time. They put the
term replication (performance) against two-phase commit
protocol.
Rule 9-11: Hardware, OS and Network Independence
User should be presented with the “single-system image”
regardless any particular hardware platform, operating system
and network.
•It is desirable to be able to run the same DBMS on
different hardware systems which all participate as
equal partners (where appropriate) in a distr. system.
The strict homogeneity assumption is not relaxed; it is still
assumed that the same DBMS is running on all those different
hardware systems.
•From a commercial point of view, the most important
operating system environments, and hence the ones that
(at a minimum) the DBMS should support, are
probably MVS/XA, MVS/ESA, VM/CMS, VAX/VMS,
UNIX (various flavours), OS/2, MS-DOS, Windows
•From the D-DBMS viewpoint, the network is merely
the provider of a reliable message transmission service.
By “reliable” here is meant that, if the network accepts a
message from site X for delivery to site Y, then it will
eventually deliver that message to site Y.
Messages will not be garbled, will not be delivered more than
once and will be delivered in the order sent.
The network should also be responsible for site
authentication.
Rule 12: DBMS independence
Ideal distributed system should provide DVBMS independence
(or transparency).
•The obvious fact is that not only the real-world
computer installations typically run on many different
machines and with many different operating systems,
but very often run with different DBMSs as well.
•All that is really needed is that the DBMS instances at
different sites all support the same interface.
•Consider an example: Suppose that site X is running
INGRES and site Y is running ORACLE. Some user at
site X desires to see a single distributed database that
includes data from INGRES (site X) and ORACLE
from site Y. By definition user U is a an INGRES user
and the distributed database must therefore be an
INGRES database.
•The solution is quite straightforward: INGRES must
provide an application program (gateway) that runs
upon ORACLE and has the effect of 'making ORACLE
look like INGRES'.
•Implementing protocols for the exchange of information
between INGRES and ORACLE involves first the
understanding of the messages in which SQL sources
are sent form INGRES and translated to ORACLE
statements and mapping ORACLE results (data values,
return codes, ...) into the messages format that INGRES
expects.
Types of parallelism in database systems
•Interquery parallelism
Multiple queries generated by concurrent transactions
can be executed parallel.
•Interoperation parallelism
A query may consist of multiple, independent
operations that can be executed parallel. There are two
forms of interoperation parallelism:
•Independent
•Pipeline
•Intraoperation parallelism
An operator may be executed as many small,
independent sub-operations. The relational model is
ideal for implementing intraoperator parallelism.
This is the most important type of parallelism, since
generally there are much more data tuples than
relational operations.
Intraoperation parallelism comes from data partitioning.
Types of parallelism in database systems
Select Select
Select Select
Inter-operator
Intra-operator
pipeline

Weitere ähnliche Inhalte

Was ist angesagt?

Intro to Distributed Database Management System
Intro to Distributed Database Management SystemIntro to Distributed Database Management System
Intro to Distributed Database Management SystemAli Raza
 
Introduction to distributed database
Introduction to distributed databaseIntroduction to distributed database
Introduction to distributed databaseSonia Panesar
 
Lecture 11 - distributed database
Lecture 11 - distributed databaseLecture 11 - distributed database
Lecture 11 - distributed databaseHoneySah
 
Distributed Database Management System(DDMS)
Distributed Database Management System(DDMS)Distributed Database Management System(DDMS)
Distributed Database Management System(DDMS)mobeen.laws
 
Distributed database management system
Distributed database management  systemDistributed database management  system
Distributed database management systemPooja Dixit
 
Distributed databases,types of database
Distributed databases,types of databaseDistributed databases,types of database
Distributed databases,types of databaseBoomadevi Shanmugam
 
Components of ddbms
Components of ddbmsComponents of ddbms
Components of ddbmsPooja Dixit
 
Distributed Database
Distributed DatabaseDistributed Database
Distributed DatabaseJovyLee4
 
Difference between Homogeneous and Heterogeneous
Difference between Homogeneous  and    HeterogeneousDifference between Homogeneous  and    Heterogeneous
Difference between Homogeneous and HeterogeneousFaraz Qaisrani
 
Types and Functions of DDBMS
Types and Functions of DDBMSTypes and Functions of DDBMS
Types and Functions of DDBMSAdeel Rasheed
 
Distributed database system
Distributed database systemDistributed database system
Distributed database systemM. Ahmad Mahmood
 
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUES
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUESDISTRIBUTED DATABASE WITH RECOVERY TECHNIQUES
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUESAAKANKSHA JAIN
 
Distributed databases and dbm ss
Distributed databases and dbm ssDistributed databases and dbm ss
Distributed databases and dbm ssMohd Arif
 

Was ist angesagt? (20)

Lecture 1 ddbms
Lecture 1 ddbmsLecture 1 ddbms
Lecture 1 ddbms
 
Intro to Distributed Database Management System
Intro to Distributed Database Management SystemIntro to Distributed Database Management System
Intro to Distributed Database Management System
 
distributed dbms
distributed dbmsdistributed dbms
distributed dbms
 
Introduction to distributed database
Introduction to distributed databaseIntroduction to distributed database
Introduction to distributed database
 
Lecture 11 - distributed database
Lecture 11 - distributed databaseLecture 11 - distributed database
Lecture 11 - distributed database
 
RDBMS
RDBMSRDBMS
RDBMS
 
Distributed Database Management System(DDMS)
Distributed Database Management System(DDMS)Distributed Database Management System(DDMS)
Distributed Database Management System(DDMS)
 
Distributed database management system
Distributed database management  systemDistributed database management  system
Distributed database management system
 
Distributed databases,types of database
Distributed databases,types of databaseDistributed databases,types of database
Distributed databases,types of database
 
Components of ddbms
Components of ddbmsComponents of ddbms
Components of ddbms
 
Distributed Database
Distributed DatabaseDistributed Database
Distributed Database
 
Difference between Homogeneous and Heterogeneous
Difference between Homogeneous  and    HeterogeneousDifference between Homogeneous  and    Heterogeneous
Difference between Homogeneous and Heterogeneous
 
Distributed database
Distributed databaseDistributed database
Distributed database
 
Types and Functions of DDBMS
Types and Functions of DDBMSTypes and Functions of DDBMS
Types and Functions of DDBMS
 
2 ddb architecture
2 ddb architecture2 ddb architecture
2 ddb architecture
 
Distributed database system
Distributed database systemDistributed database system
Distributed database system
 
Distributed dbms
Distributed dbmsDistributed dbms
Distributed dbms
 
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUES
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUESDISTRIBUTED DATABASE WITH RECOVERY TECHNIQUES
DISTRIBUTED DATABASE WITH RECOVERY TECHNIQUES
 
Distributed databases and dbm ss
Distributed databases and dbm ssDistributed databases and dbm ss
Distributed databases and dbm ss
 
Distributed DBMS - Unit 3 - Distributed DBMS Architecture
Distributed DBMS - Unit 3 - Distributed DBMS ArchitectureDistributed DBMS - Unit 3 - Distributed DBMS Architecture
Distributed DBMS - Unit 3 - Distributed DBMS Architecture
 

Ähnlich wie Csld phan tan va song song

lecture-13.pptx
lecture-13.pptxlecture-13.pptx
lecture-13.pptxlaiba29012
 
Adbms 31 data distribution transparency
Adbms 31 data distribution transparencyAdbms 31 data distribution transparency
Adbms 31 data distribution transparencyVaibhav Khanna
 
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptx
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptxSantosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptx
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptxSANTOSH KUMAR MEHER
 
Distributed Database Management System.pptx
Distributed Database Management System.pptxDistributed Database Management System.pptx
Distributed Database Management System.pptxMajidRajper4
 
Architecture Design in Software Engineering
Architecture Design in Software EngineeringArchitecture Design in Software Engineering
Architecture Design in Software Engineeringcricket2ime
 
Chapter 4 security part ii auditing database systems
Chapter 4 security part ii auditing database systemsChapter 4 security part ii auditing database systems
Chapter 4 security part ii auditing database systemsjayussuryawan
 
Distributed database management system
Distributed database management systemDistributed database management system
Distributed database management systemVinay D. Patel
 
Shared information systems
Shared information systemsShared information systems
Shared information systemsHimanshu
 
Distributed database management systems
Distributed database management systemsDistributed database management systems
Distributed database management systemsDhani Ahmad
 
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...Big Data Storage Concepts from the "Big Data concepts Technology and Architec...
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...raghdooosh
 
Database management system
Database management systemDatabase management system
Database management systemAmit Sarkar
 
Advance database system (part 2)
Advance database system (part 2)Advance database system (part 2)
Advance database system (part 2)Abdullah Khosa
 
10 architectural design
10 architectural design10 architectural design
10 architectural designAyesha Bhatti
 
10 architectural design (1)
10 architectural design (1)10 architectural design (1)
10 architectural design (1)Ayesha Bhatti
 
Distributed database. pdf
Distributed database. pdfDistributed database. pdf
Distributed database. pdfSurajGhadge15
 

Ähnlich wie Csld phan tan va song song (20)

lecture-13.pptx
lecture-13.pptxlecture-13.pptx
lecture-13.pptx
 
Adbms 31 data distribution transparency
Adbms 31 data distribution transparencyAdbms 31 data distribution transparency
Adbms 31 data distribution transparency
 
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptx
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptxSantosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptx
Santosh Kumar Meher(2105040008) DISTRIBUTED DATABASE.pptx
 
Distributed Database Management System.pptx
Distributed Database Management System.pptxDistributed Database Management System.pptx
Distributed Database Management System.pptx
 
Architecture Design in Software Engineering
Architecture Design in Software EngineeringArchitecture Design in Software Engineering
Architecture Design in Software Engineering
 
Chapter 4 security part ii auditing database systems
Chapter 4 security part ii auditing database systemsChapter 4 security part ii auditing database systems
Chapter 4 security part ii auditing database systems
 
Distributed dbms (ddbms)
Distributed dbms (ddbms)Distributed dbms (ddbms)
Distributed dbms (ddbms)
 
Distributed database management system
Distributed database management systemDistributed database management system
Distributed database management system
 
unit 1.pdf
unit 1.pdfunit 1.pdf
unit 1.pdf
 
Shared information systems
Shared information systemsShared information systems
Shared information systems
 
distributed system original.pdf
distributed system original.pdfdistributed system original.pdf
distributed system original.pdf
 
Distributed database management systems
Distributed database management systemsDistributed database management systems
Distributed database management systems
 
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...Big Data Storage Concepts from the "Big Data concepts Technology and Architec...
Big Data Storage Concepts from the "Big Data concepts Technology and Architec...
 
Database management system
Database management systemDatabase management system
Database management system
 
Advance database system (part 2)
Advance database system (part 2)Advance database system (part 2)
Advance database system (part 2)
 
DBMS.pptx
DBMS.pptxDBMS.pptx
DBMS.pptx
 
10 architectural design
10 architectural design10 architectural design
10 architectural design
 
10 architectural design (1)
10 architectural design (1)10 architectural design (1)
10 architectural design (1)
 
Unit 1 dbms
Unit 1 dbmsUnit 1 dbms
Unit 1 dbms
 
Distributed database. pdf
Distributed database. pdfDistributed database. pdf
Distributed database. pdf
 

Kürzlich hochgeladen

VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our EscortsVIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escortssonatiwari757
 
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...ranjana rawat
 
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
 
2024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 302024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 30JSchaus & Associates
 
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our EscortsVIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escortssonatiwari757
 
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
 
Night 7k to 12k Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...
Night 7k to 12k  Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...Night 7k to 12k  Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...
Night 7k to 12k Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...aartirawatdelhi
 
2024 Zoom Reinstein Legacy Asbestos Webinar
2024 Zoom Reinstein Legacy Asbestos Webinar2024 Zoom Reinstein Legacy Asbestos Webinar
2024 Zoom Reinstein Legacy Asbestos WebinarLinda Reinstein
 
The U.S. Budget and Economic Outlook (Presentation)
The U.S. Budget and Economic Outlook (Presentation)The U.S. Budget and Economic Outlook (Presentation)
The U.S. Budget and Economic Outlook (Presentation)Congressional Budget Office
 
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...SUHANI PANDEY
 
Election 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdfElection 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdfSamirsinh Parmar
 
EDUROOT SME_ Performance upto March-2024.pptx
EDUROOT SME_ Performance upto March-2024.pptxEDUROOT SME_ Performance upto March-2024.pptx
EDUROOT SME_ Performance upto March-2024.pptxaaryamanorathofficia
 
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...anilsa9823
 
Zechariah Boodey Farmstead Collaborative presentation - Humble Beginnings
Zechariah Boodey Farmstead Collaborative presentation -  Humble BeginningsZechariah Boodey Farmstead Collaborative presentation -  Humble Beginnings
Zechariah Boodey Farmstead Collaborative presentation - Humble Beginningsinfo695895
 
Human-AI Collaboration for Virtual Capacity in Emergency Operation Centers (E...
Human-AI Collaborationfor Virtual Capacity in Emergency Operation Centers (E...Human-AI Collaborationfor Virtual Capacity in Emergency Operation Centers (E...
Human-AI Collaboration for Virtual Capacity in Emergency Operation Centers (E...Hemant Purohit
 
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...tanu pandey
 
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Bookingdharasingh5698
 
Top Rated Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...
Top Rated  Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...Top Rated  Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...
Top Rated Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...Call Girls in Nagpur High Profile
 

Kürzlich hochgeladen (20)

VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our EscortsVIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl mohali 7001035870 Enjoy Call Girls With Our Escorts
 
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...
↑VVIP celebrity ( Pune ) Serampore Call Girls 8250192130 unlimited shot and a...
 
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
 
2024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 302024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 30
 
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our EscortsVIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escorts
VIP Call Girl Service Ludhiana 7001035870 Enjoy Call Girls With Our Escorts
 
Rohini Sector 37 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 37 Call Girls Delhi 9999965857 @Sabina Saikh No AdvanceRohini Sector 37 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
Rohini Sector 37 Call Girls Delhi 9999965857 @Sabina Saikh No Advance
 
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Nanded City Call Me 7737669865 Budget Friendly No Advance Booking
 
Night 7k to 12k Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...
Night 7k to 12k  Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...Night 7k to 12k  Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...
Night 7k to 12k Call Girls Service In Navi Mumbai 👉 BOOK NOW 9833363713 👈 ♀️...
 
2024 Zoom Reinstein Legacy Asbestos Webinar
2024 Zoom Reinstein Legacy Asbestos Webinar2024 Zoom Reinstein Legacy Asbestos Webinar
2024 Zoom Reinstein Legacy Asbestos Webinar
 
The U.S. Budget and Economic Outlook (Presentation)
The U.S. Budget and Economic Outlook (Presentation)The U.S. Budget and Economic Outlook (Presentation)
The U.S. Budget and Economic Outlook (Presentation)
 
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...
VIP Model Call Girls Kiwale ( Pune ) Call ON 8005736733 Starting From 5K to 2...
 
Election 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdfElection 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdf
 
EDUROOT SME_ Performance upto March-2024.pptx
EDUROOT SME_ Performance upto March-2024.pptxEDUROOT SME_ Performance upto March-2024.pptx
EDUROOT SME_ Performance upto March-2024.pptx
 
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...
Lucknow 💋 Russian Call Girls Lucknow ₹7.5k Pick Up & Drop With Cash Payment 8...
 
Zechariah Boodey Farmstead Collaborative presentation - Humble Beginnings
Zechariah Boodey Farmstead Collaborative presentation -  Humble BeginningsZechariah Boodey Farmstead Collaborative presentation -  Humble Beginnings
Zechariah Boodey Farmstead Collaborative presentation - Humble Beginnings
 
Human-AI Collaboration for Virtual Capacity in Emergency Operation Centers (E...
Human-AI Collaborationfor Virtual Capacity in Emergency Operation Centers (E...Human-AI Collaborationfor Virtual Capacity in Emergency Operation Centers (E...
Human-AI Collaboration for Virtual Capacity in Emergency Operation Centers (E...
 
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
 
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Bhavnagar 7001035870 Whatsapp Number, 24/07 Booking
 
Top Rated Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...
Top Rated  Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...Top Rated  Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...
Top Rated Pune Call Girls Hadapsar ⟟ 6297143586 ⟟ Call Me For Genuine Sex Se...
 
(NEHA) Call Girls Nagpur Call Now 8250077686 Nagpur Escorts 24x7
(NEHA) Call Girls Nagpur Call Now 8250077686 Nagpur Escorts 24x7(NEHA) Call Girls Nagpur Call Now 8250077686 Nagpur Escorts 24x7
(NEHA) Call Girls Nagpur Call Now 8250077686 Nagpur Escorts 24x7
 

Csld phan tan va song song

  • 1. Concepts of Parallel and Distributed Database Systems Key concepts: - concept and structure of parallel and distributed databases - motivations for parallelism and distribution - types of parallelism - benefits of DDBMS - drawbacks of DDBMS - Date’s rules - local autonomy - function distribution - continues operation - location independence - fragmentation independence - replication independence - overview of query processing - overview of distributed transaction management - hardware independence - DBMS independence - Distribution transparency - Replication transparency - Fragmentation transparency - history of parallel databases - history of distributed databases
  • 2. Parallel and Distributed Databases •A parallel database aims principally linear speedup and scaleup: Linear scaleup refers to a sustained performance for a linear increase both in database size and processing and storage power. Linear speedup refers to a linear increase in performance for a constant database size. •Sharing data is the key of a distributed DB → cooperative approach Since data is distributed, users that share that data can have it placed at the site they work on, with local control (Local autonomy) •Distributed and parallel databases improve reliability and availability I.e. they have replicated components and thus eliminate single points of failure. •The distribution of data and the parallel/distributed processing is not visible to the users → transparency •Distributed Database (DDB)
  • 3. A collection of multiple, logically interrelated databases distributed over a computer network •Distributed Database Management System (D-DBMS) The software system that permits the management of the DDB and provides an access mechanism that makes this distribution transparent to the users. •Distributed Database System (DDBS) DDBS = DDB + D-DBMS •What is not a DDBS: A database system which resides on •a timesharing computer system •a loosely or tightly coupled multiprocessor system •one of the nodes of a network of computers - this is a centralised database on a network node Reasons of data distribution •Improved reliability and availability through distributed transactions •Improved performance
  • 4. •Allowing data sharing while maintaining some measure of local control •Easier and more economical system expansion •Distributed nature of some database applications Additional functionality of a D-DBMS •Distribution leads to increased complexity in the system design and implementation •D-DBMS must be able to provide some additional functionality to those of a central DBMS: • To access remote sites and transmit queries and data among the various sites via a communication network. • To keep track of the data distribution and replication in the DDBMS catalog. • To devise execution strategies for queries and transactions that access data from more than one site. • To decide on which copy of a replicated data item to access. • To maintain the consistency of copies of a replicated data item. • To maintain the global conceptual schema of the distributed database. • To recover from individual site crashes and from new types of failures such as failure of a communication link. Distributed DBMS Issues •Distributed Database Design
  • 5. •how to distribute the database •replicated & non-replicated database distribution •a related problem in directory management • Query Processing •convert user transactions to data manipulation instructions •optimisation problem •min{cost = data transmission + local processing} •general formulation is NP-hard •Concurrency Control •synchronisation of concurrent accesses •consistency and isolation of transactions' effects •deadlock management •Reliability •how to make the system resilient to failures •atomicity and durability
  • 6. Date’s 12 rules for distributed database systems Rule 0: to the user, a distributed system should look exactly like a non-distributed system other rules: 1. Local autonomy 2. No reliance on central site 3. Continuous operation 4. Location independence 5. Fragmentation independence 6. Replication independence 7. Distributed query processing 8. Distributed transaction management 9. Hardware independence 10. Operating system independence 11. Network independence 12. DBMS independence The great principles are:
  • 7. Autonomy, Independence and Transparency. Rule 1: Local Autonomy Autonomy objective: Sites should be autonomous to the maximum extent possible. •Local data is locally owned and managed, with local accountability •security considerations •integrity considerations •Local operations remain purely local •All operations at a given site are controlled by that site; no site X should depend on some other site Y for its successful functioning Otherwise, if site Y is down, site X could not carry out the operation although there is nothing wrong at site X. •In some situations some slight loss of autonomy is inevitable •fragmentation problem – rule 5 •replication problem – rule 6 •update of replicated relation – rule 6 •multiple-site integrity constraint problem – rule 7 •a problem of participation in a two-phase commit process – rule 8
  • 8. In consequence: All data ‘really’ belongs to some local database even it is accessible from remote. Rule 2: No reliance on central site There must not be any reliance on a central ‘master’site for some central service, such as centralised query processing or centralised transaction management, such that the entire system is dependent on that central site. •Reliance on a central site would be undesirable for at least the following two reasons: •that central site might be a bottleneck •the system would be vulnerable •In a distributed system therefore, the following functions (among others) must all be distributed: •Dictionary management •Query processing •Concurrency control •Recovery control Rule 3: Continuous operation
  • 9. There should ideally never be any need for a planned entire system shutdown . •Incorporating a new site into an existing distributed system should not bring the entire system to a halt •Incorporating a new site into an existing distributed system should not require any changes to existing user programs or terminal activities •Removing an existing site from the distributed systems should not cause any unnecessary interruptions in service •Within the distributed system, it should be possible to create and destroy fragments and replicas of fragments dynamically •It should be possible to upgrade the DBMS at any given component site to a newer release without taking the entire system down • Rule 4: Location independence (Transparency) Users should not have to know where data is physically stored but rather should be able to behave – at least from a logical standpoint – as if the data was all stored at their own local site.
  • 10. •Sometimes referenced as Transparency •Simplifies user programs and terminal activities •Allows data to migrate from site to site •It is easier to provide location independence for simple retrieval operations than it is for update operations •Distributed data naming scheme and corresponding support from the dictionary subsystem •User naming scheme •An user has to have a valid logon ID at each of multiple sites to operate •User profile for each valid logon ID in the dictionary •Granting of access privileges at each component site • Rule 5: Fragmentation independence (Transparency) •A distributed system supports data fragmentation if a given relation can be divided up into pieces or ‘fragments’for physical storage purposes •A system that supports data fragmentation should also support fragmentation independence (also known as fragmentation transparency)
  • 11. Users should be able to behave (at least from a logical standpoint) as if the data were in fact not fragmented at all. •Fragmentation is desirable for performance reasons •Fragmentation must be defined within the context of a distributed database •Fragmentation independence (like location independence) is desirable because it simplifies user programs and terminal activities •Fragmentation independence implies that users should normally be presented with a view of the data in which the fragments are logically combined together by means of suitable joins and unions Fragmentation principles •Two basic kinds of fragmentation •Horizontal fragmentation • refers to cut between tuples • achieved by selection • same as data partitioning in parallel databases • efficient for parallel scanning of the relation •Vertical fragmentation • refers to the cut of the schema
  • 12. • achieved by a projection • efficient if there is frequent access to different attribute groups •The union of the horizontal fragments should be the original relation •The join of the vertical fragments should be the original relation !!! Supplemental measure must be undertaken to guarantee that the join is nonloss! E.g. the fragmentation of the table Employee( Emp#, Dept#, Salary) into the fragments Emp#, Dept# and Salary would not be valid, since there is no common join-attribute. Solutions are either to duplicate the key or to introduce a hidden ‘tupleID’ where the TID is the physical or logical address for that tuple. This TID is then included in all fragments. Rule 6: Replication independence (Transparency) User should be able to behave as if the data were in fact not replicated at all.
  • 13. •A distributed system supports data replication if a given relation (more generally, a given fragments of a relation) can be represented at the physical level by many distinct stored copies or replicas, at many distinct sites. •Replication, like fragmentation, should be “transparent to the user” •Replication is desirable for at least two reasons: •Performance: applications can operate on local copies instead of having communicate with remote sites (or a ‘nearest’copy can be fetched. •Availability: a given replicated object remains available for processing as long as at least one copy remains available. •Update propagation problem! •Fragmentation independence (like all transparency rules) is desirable because it simplifies user programs and terminal activities Update propagation problem •The major disadvantage of replication is, of course, when a given replicated object is updated, all copies of that object must be updated. → Update propagation problem.
  • 14. •A common scheme for dealing with this problem is the so-called primary copy scheme: • One copy of each replicated object is designated as the primary copy. The remainder are all secondary copies. • Primary copies of different objects are at different sites (so the distributed scheme once again). • Update operations are deemed logically complete as soon as the primary copy has been update. The site holding the copy is then responsible for propagating the update to the secondary copies at some subsequent time. •The 'subsequent' time must be prior to COMMIT, however, if the ACID properties of the transaction are to be preserved → Inconsistency ?! (see below: transaction processing). Rule 7: Distributed query processing It is crucially important for distributed database systems to choose a good strategy for distributed query processing. •Query processing in a distributed system involve •local CPU and I/O activity at several distinct sites •some amount of data communication among those sites
  • 15. •Amount of data communication is a major performance factor. Depending on the bandwidth and the relation sizes the minimal communication cost strategy will be chosen. •Distributed query processing and query optimisation must be aware of the distributed context: •Where is a relation (fragment) stored? •Are there any replicas? •Query compilation ahead of time •Views that span multiple sites •Integrity constraints within a DDBS that span multiple sites Rule 8: Distributed transaction management Two major aspects of transaction management, recovery control and concurrency control, require extended treatment in the distributed environment •In a distributed system, a single transaction can involve the execution of code at multiple sites and can thus involve updates at multiple sites
  • 16. •Each transaction is therefore said to consist of multiple “agents”, where an agent is the process performed on behalf of a given transaction at a given site •Recovery Control: In order to ensure that a given transaction is atomic in the distributed environment, therefore, the system must ensure that the set of agents for that transaction either all commit in unison or all roll back in unison. That effect can be achieved by means of the two-phase commit protocol. •Concurrency Control: is based in most distributed systems on locking (some implemented multi-version control ..., read can have the last version) Recovery control: Two-phase commit protocol •In general, the two-phase commit protocol is driven, whenever a transaction interact with several independent 'resource manager', i.e. each managing its own set of recoverable resources and maintaining its own recovery log. •FIRST PHASE: Each site must force-write all log entries for local resources used by the transaction out to its physical log. Assuming the force-write is successful, each site gives the OK.
  • 17. •SECOND PHASE: If the site where the transaction has been submitted receives all the OK form the other sites, it force-writes an entry to its own physical log. If anything has gone noOK, then general ROLLBACK otherwise the decision is COMMIT DONE. •Remark 1: If site Y acts as a participant in a two-phase commit coordinated by site X, then Y must do what is told by site X (roll-back) – a (minor) loss of local autonomy. •Remark 2: If one of the messages get lost (e.g. ROLLBACK) then the system is an inconsistent state. In general there can be no protocol that can guarantee that all agents will commit unison. Concurrency control: Primary Copy approach •Lock means: Requests to test, set and release locks. Update means update and acknowledgement. •Consider for example a transaction that needs to update an object for which exists replicas at n sites. A straightforward implementation will require: n lock requests; n lock grants; n update messages; n acknowledgements; n unlock requests makes 5*n messages (could of course be piggy-picked)
  • 18. •Use the primary copy: Lock only the primary copy (one lock request, one lock grant and one unlock) and the site with the primary copy handles the updating (2*n messages). Thus we got only 2*n+3 messages. •Remark 1: Loss of autonomy: A transaction can fail now if a primary copy is unavailable, and a secondary copy exists at the local site. •Remark 2: Inconsistency is introduced if all secondary copies are not updated prior to the COMMIT point. Unfortunately, many commercial products support a less ambitious form of replication in which the update propagation is guaranteed to be done at some future time. They put the term replication (performance) against two-phase commit protocol. Rule 9-11: Hardware, OS and Network Independence User should be presented with the “single-system image” regardless any particular hardware platform, operating system and network. •It is desirable to be able to run the same DBMS on different hardware systems which all participate as equal partners (where appropriate) in a distr. system.
  • 19. The strict homogeneity assumption is not relaxed; it is still assumed that the same DBMS is running on all those different hardware systems. •From a commercial point of view, the most important operating system environments, and hence the ones that (at a minimum) the DBMS should support, are probably MVS/XA, MVS/ESA, VM/CMS, VAX/VMS, UNIX (various flavours), OS/2, MS-DOS, Windows •From the D-DBMS viewpoint, the network is merely the provider of a reliable message transmission service. By “reliable” here is meant that, if the network accepts a message from site X for delivery to site Y, then it will eventually deliver that message to site Y. Messages will not be garbled, will not be delivered more than once and will be delivered in the order sent. The network should also be responsible for site authentication. Rule 12: DBMS independence Ideal distributed system should provide DVBMS independence (or transparency).
  • 20. •The obvious fact is that not only the real-world computer installations typically run on many different machines and with many different operating systems, but very often run with different DBMSs as well. •All that is really needed is that the DBMS instances at different sites all support the same interface. •Consider an example: Suppose that site X is running INGRES and site Y is running ORACLE. Some user at site X desires to see a single distributed database that includes data from INGRES (site X) and ORACLE from site Y. By definition user U is a an INGRES user and the distributed database must therefore be an INGRES database. •The solution is quite straightforward: INGRES must provide an application program (gateway) that runs upon ORACLE and has the effect of 'making ORACLE look like INGRES'. •Implementing protocols for the exchange of information between INGRES and ORACLE involves first the understanding of the messages in which SQL sources are sent form INGRES and translated to ORACLE statements and mapping ORACLE results (data values, return codes, ...) into the messages format that INGRES expects.
  • 21. Types of parallelism in database systems •Interquery parallelism Multiple queries generated by concurrent transactions can be executed parallel. •Interoperation parallelism A query may consist of multiple, independent operations that can be executed parallel. There are two forms of interoperation parallelism: •Independent •Pipeline •Intraoperation parallelism An operator may be executed as many small, independent sub-operations. The relational model is ideal for implementing intraoperator parallelism. This is the most important type of parallelism, since generally there are much more data tuples than relational operations. Intraoperation parallelism comes from data partitioning.
  • 22. Types of parallelism in database systems Select Select Select Select Inter-operator Intra-operator pipeline