Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
2Activity
0 of .
Results for:
No results containing your search query
P. 1
Concurrency Control In CAD Using Functional Back Propagation Neural Network

Concurrency Control In CAD Using Functional Back Propagation Neural Network

Ratings: (0)|Views: 21 |Likes:
Published by ijcsis
This paper presents artificial neural network method using functional back propagation algorithm (FUBPA) for implementing concurrency Control while developing dial of a fork using Autodesk inventor 2008. Initially, the various parts are decided and the sequence in which they have to be drawn. While implementing concurrency control, this work ensures that associated parts cannot be accessed by more than one person due to locking. The FUBPA learns the objects and the type of transactions to be done based on which node in the output layer of the network exceeds a threshold value. Learning stops once all the objects are exposed to FUBPA. During testing performance, metrices are analyzed.
This paper presents artificial neural network method using functional back propagation algorithm (FUBPA) for implementing concurrency Control while developing dial of a fork using Autodesk inventor 2008. Initially, the various parts are decided and the sequence in which they have to be drawn. While implementing concurrency control, this work ensures that associated parts cannot be accessed by more than one person due to locking. The FUBPA learns the objects and the type of transactions to be done based on which node in the output layer of the network exceeds a threshold value. Learning stops once all the objects are exposed to FUBPA. During testing performance, metrices are analyzed.

More info:

Published by: ijcsis on Mar 08, 2011
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

03/19/2011

pdf

text

original

 
 
CONCURRENCY CONTROL IN CADUSING FUNCTIONAL BACKPROPAGATION NEURAL NETWORK
A.Muthukumaravel, Dr.S.Purushothaman and Dr.A.Jothi
A.Muthukumaravel
Research Scholar
Department of MCA
Vels university
Chennai-600117, India
Dr.S.Purushothaman
Principal,Sun College of Engineering andTechnology,Sun Nagar, Erachakulum,Kanyakumari District-629902, India
Dr.A.jothiDean,School of Computing SciencesVels universityChennai-600117, India
ABSTRACT 
--
This paper presents artificial neuralnetwork method using functional back propagationalgorithm (FUBPA) for implementing concurrencyControl while developing dial of a fork using Autodeskinventor 2008. Initially, the various parts are decidedand the sequence in which they have to be drawn. Whileimplementing concurrency control, this work ensuresthat associated parts cannot be accessed by more thanone person due to locking. The FUBPA learns theobjects and the type of transactions to be done based onwhich node in the output layer of the network exceeds athreshold value. Learning stops once all the objects areexposed to FUBPA. During testing performance,metrices are analyzed.
Keywords:
 
Concurrency Control, Functional Back Propagation
Network, Tr
ansaction Locks, TimeStamping.
I.
 
INTRODUCTIONMaintaining consistency in transactions of objects during drawing huge computer aided object isthe result of efficient concurrency Control. Incomputer aided design (CAD), many persons will beaccessing different parts of same objects according tothe type work allotted to engineers. As all the parts of the same objects are stored in a single file, at anypoint of time, there should not be corruption of data,inconsistency in storage and total loss of data.Locks are used for accessing objects. In adatabase operation lock manager plays an importantrole whether one or more transactions are reading orwriting any part of ‘I’ where ‘I’ is an item. It is thepart of that record, for each item I. Gaining access toI is controlled by manager and ensure that there is no, access (read or write) would cause a conflict. The
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 2, February 2011168 http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
 
lock manager can store the current locks in a lock table which consists of records with fields (<object>,<lock type>, <transaction>) the meaning of record (I,L, T) is that transaction T has a lock of type L onobject I [1-4].The process of managing simultaneousoperations on the database without having theminterfere with one another is called concurrency.[5-8]When two or more users are accessing databasesimultaneously concurrency prevents interference..Interleaving of operations may produce an incorrectresult even though two transactions may be correct.Some of the problems that result in concurrency arelost update, inconsistent analysis and uncommitteddependency.II.
 
PROBLEM DEFINITIONThere is inability to provide consistency inthe database when long transactions are involved. Itwill not be able to identify if there is any violation of database consistency during the time of commitment.It is not possible to know, if the transaction is withundefined time limit. There is no serializability whenmany users work on shared objects. During longtransactions, optimistic transactions and two phaselocking will result in deadlock. Two phase lockingforces to lock resources for long time even after theyhave finished using them. Other transactions thatneed to access the same resources are blocked. Theproblem in optimistic mechanism with TimeStamping is that it causes repeated rollback of transactions when the rate of conflicts increasessignificantly.
 
Artificial neural network [9] withFunctional Back Propagation Network (FUBPA) hasbeen used to manage the locks allotted to objects andlocks are claimed
 
appropriately to be allotted forother objects during subsequent transactions.Inbuilt library drawing for the dial of fork (Figure 1)are available in AutoCAD. The fork is used in thetwo wheeler front structure. Due to customerrequirements, the designer edits the dial of fork in thecentral database by modifying different features.Consistency of the data has to be maintained duringthe process of modifications of different features.Following sequences of locking objects have to bedone whenever a particular user accesses a specificfeature of the dial of fork. Each feature is treated asan object. The features are identified with numbersand corresponding feature names. In this explanation,O
1
refers to an object / feature marked as 1.In general, the following sequences are formed whencreating dial of fork. The major parameters involvedin creating the dial of the fork are hollow cylinder,wedge and swiveling plate. The various constraintsthat have to be imposed during modifications of features by many users on this dial of fork are asfollows:
 
During development of features, hollowcylinder details should not be changed
 
External rings are associated with hollowcylinder.
 
The circular wedge has specific slope andassociated with hollow cylinder.This dial of fork has following entities.1) Features 1, 2, (set 1)2) Features 10, 11,12,13,14 (set 2)3) Features 5,6,7,8 (set 3)4) Features 3, 4 (set 4)
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 2, February 2011169 http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
 
Figure 1.
 
Dial of fork 
1 Lower end, 2 height of the end part, 3 external support, 4 Height of the externalsupport, 5 support for the wedge, 6 height of the support for the wedge, 7 wedge, 8thickness of the wedge, 9 slope of the wedge, 10 Wedge lock, 11 Height of the wedgelock, 12 Concentric hole, 13 separator, 14 Guideway
Set 1, set 2, set 3, set 4 can be made intoindividual drawing part files (part file 1, part file 2,part file 3 and part file 4) and combined into oneassembly file (containing the part files 1,2 3 and 4which will be intact). When the users are accessingindividual part files, then transactions in part file 1need not worry about the type of transactions in partfiles 2,3,4 and vice versa among them. When the partfiles 1 2, 3 and 4 are combined into a single assemblyfile, then inconsistency in the shape and dimension of the set 1, set 2, set 3 and set 4, during matchingshould not occur. Provisions can be made incontrolling the dimensions and shapes with upper andlower limits confirming to standards. At any timewhen a subsequent user is trying to access lockedfeatures, he can modify the features on his systemand store as an additional modified copy of thefeatures with Time Stamping and version names(allotted by the user / allotted by the system).
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 2, February 2011170 http://sites.google.com/site/ijcsis/ISSN 1947-5500

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->