SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
Unit-3
Data management issues in mobile computing
Data management that allows to access data from devices is the main problem.
The mobile database allows applications to access data from devices even when
on the move. It is a database that connects devices over a wireless network. Here,
the client and server have a wireless connection. The wired connections have the
entire database distributed with full or partial replication. Fixed hosts are there to
perform data management with the help of database servers. Base stations are
present at fixed locations to pass communication with the mobile units to and from
the fixed hosts.
Issues that arise in data management of mobile databases are:
Mobile Database Design
On account of the continuous closure and for dealing with the queries, the
worldwide name goal issue is compounded.
Security
Data Distribution and Replication
Replication Issues
Division of labor
Transaction Models
Recovery and Fault Tolerance
Location? Based Service
Query Processing
Faculty Signature
SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
Data replication for mobile computers
Data replication in mobile computing involves creating and storing multiple
copies of data across different locations, including on the mobile device itself and
on a central server, to ensure data availability, consistency, and fault
tolerance. This is crucial because mobile devices often experience intermittent
connectivity and limited resources, making it essential to access data even offline
and to minimize data loss or corruption.
Here's a more detailed explanation:
Why Data Replication in Mobile Computing?
Availability:
Mobile users need access to their data regardless of network
conditions. Replication allows users to access data even when offline, and data can
be restored from a backup when connectivity is restored.
Reliability and Fault Tolerance:
Multiple copies of data reduce the risk of data loss due to hardware failures,
software glitches, or network outages. If one copy is unavailable, another can be
accessed.
Reduced Communication Costs:
By storing data locally, mobile devices can reduce the need to constantly
communicate with the central server, saving bandwidth and battery life.
Data Sharing:
Data replication facilitates data sharing between multiple users or devices by
providing consistent copies across different locations.
Security:
Faculty Signature
SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
Replication can enhance security by creating backup copies of sensitive data,
making it more resilient to data breaches or accidental deletions.
Types of Replications:
Full Replication: All data is copied to multiple locations.
Partial Replication: Only specific fragments of data are replicated.
Synchronous Replication: Changes to the original data are immediately
replicated to all copies.
Asynchronous Replication: Changes are replicated after a transaction is
committed.
Transactional Replication: Individual changes are replicated, often used
for databases.
Adaptive clustering for mobile wireless networks
Adaptive Clustering for Mobile Wireless Networks:
Clustering- It are often considered the foremost important unsupervised learning
problem. It involves finding structure during a set of unlabelled data. Definition of
cluster-"The process of organizing objects into groups whose members are
somewhat similar." Therefore, a cluster could also be a set of objects that are
"similar" to them/not almost like objects belonging to other clusters.
Clustering: Two methods for clustering are-
Distance: according to a given distance, if two or more objects are close, they
belong to the same cluster. This is often called distance-based clustering.
Conceptual: If two or more objects belong to the same class, it's considered that
this is often a "common" concept for of those objects, namely the conceptual
class.
Goal: Determine the inherent grouping during a group of unlabelled data.
Requirements: clustering algorithm. Should be satisfied
Faculty Signature
SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
• Scalability
• The ability to affect noise and outliers
• High-dimensional
• Interpretability and usefulness
• Handle differing kinds of attributes
• Discover star clusters of any shape
Application: clustering algorithm. Can be used in many fields some of are-
• Marketing-customer groups with similar behaviours
• Biology-classification of animals and plants
• Library-Book Order
• Insurance – Determine the group of auto insurance policy holders
• Earthquake research-a collection of observed earthquake epicentres
• www-file classification
• Urban planning-determine the house group according to the house sign. To
their house type
Clustering Algorithm: It may be classified as –
• k-means (exclusive)
• Fuzzy C mean (overlap)
• Hierarchical clustering (hierarchical)
• Gaussian mixture (probability)
In mobile computing, a file system is the organizational structure used by an
operating system to store, manage, and access files on a storage device like a
mobile phone's memory or a microSD card. It dictates how data is stored and
accessed, ensuring efficient data organization and retrieval for mobile devices.
Here's a more detailed look:
1. Function of a File System:
Faculty Signature
SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
Organization:
File systems provide a structured way to organize files into directories and
subdirectories, making it easy to find and access specific files.
Storage:
They manage the allocation of storage space on the device, ensuring that files are
stored efficiently and don't overlap.
Access:
File systems provide the interface for applications to read, write, and delete files.
Data Management:
They handle various data management tasks like file naming conventions, file
permissions, and file integrity.
Disconnected operations
Disconnected operation in mobile computing refers to the ability of a mobile
device to continue functioning and accessing data even when its network
connection is lost. This is achieved by storing data locally on the device and
managing updates when connectivity is restored.
Key Aspects of Disconnected Operations:
Local Caching:
Data is stored locally on the mobile device, allowing access even without network
connectivity.
Synchronization:
When a connection is reestablished, the device synchronizes its local data with the
central repository, resolving any conflicts that may have arisen.
Faculty Signature
SUNDER DEEP COLLEGE OF MANAGEMENT & TECHNOLOGY,
GHAZIABAD
(Affiliated with CCSU, MEERUT)
Topic Covered_____________________________________________________
Lecture No. __________ Date & Time of Delivering______________________
Reduced Dependence on Network:
Disconnected operation minimizes reliance on a continuous network connection,
making mobile computing more robust and reliable in areas with intermittent or
unreliable network coverage.
Examples:
This concept is crucial for applications like mobile banking, offline map navigation,
and document editing, where users need to work with data even when not
connected to the internet.
Faculty Signature
