Post on 08-Apr-2018
transcript
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
1/17
Shared Data Allocation in Mobile
Computing-Using Local optimizationand Global optimization
Phanikanth.ch10102111
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
2/17
Mobile Computing
cellular
bluetooth
WiFi
UWB
satellite
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
3/17
Shared Data
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
4/17
An example scenario of shared dataallocation problem
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
5/17
In this Paper, we device data allocation algorithms that can utilize the
knowledge of user moving patterns for proper allocation of shareddata.
Performance of Mobile computing system can be improved by properallocation of shared data.
Based on moving behaviors users can be further divided into twotypes:
Frequently moving usersInfrequently moving users
Closeness measure which corresponds to the amount of the intersectionbetween the set of frequently moving user patterns and that of infrequently moving user patterns
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
6/17
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
7/17
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
8/17
Local optimization
Local optimization refers to the optimization that the likelihood of localdata access by an individual mobile user is maximized , meaning that theprobability of average local access is maximized.
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
9/17
Global optimization
Global optimization refers to the optimization that the likelihood of local dataaccess by all mobile users is maximized , meaning that the number of totallocal access is maximized.
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
10/17
Algorithm SD-local
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
11/17
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
12/17
Algorithm SD-global
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
13/17
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
14/17
Local hit ratios of mobile users using SD-local and SD-global are higherthan that using DF.
Infrequently moving users U2,U3,U4 will have better local access hitswhen using SD-local than using SD-global.Frequently moving user U1 performs better under SD-globalSD-Local favors infrequently moving users.
SD-Global favors frequently moving users.
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
15/17
T he Impact of Closeness Measure
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
16/17
Comp aris on analysis of SD-Local and SD-gl ob al
8/7/2019 Shared Data Allocation in Mobile Computing-Using Local optimiztion
17/17
Conclusion
we devised data allocation schemes that utilize the knowledge of usermoving patterns for proper allocation of shard data in a mobilecomputing system.
Algorithms SD-local and SD-global are devised to achieve localoptimization and global optimization.
SD-local favors infrequently moving users and SD-global favors frequentlymoving users.
T he knowledge obtained from the user moving patterns is very importantin devising effective shared data allocation algorithms.