Sunday, September 20, 2015
Tuesday, September 15, 2015
Bankers algorithm program
Banker's Algorithm
Goal:- C program For
Banker's Algorithm.
Method:-
Simple
Rules Of Banker Algorithm.
Explanation:-
Banker's
Algorithm is a deadlock avoidance algorithm that checks for safe or unsafe
state of a System after allocating resources to a process.
When a
new process enters into system ,it must declare maximum no. of instances
of each resource that it may need.After requesting operating system
run banker's algorithm to check whether after allocating requested
resources,system goes into deadlock state or not. If yes then it will deny the
request of resources made by process else it allocate resources to that
process.
No. of
requested resources (instances of each resource) may not exceed no. of
available resources in operating system and when a process completes it must
release all the requested and already allocated resources.
For
implementing Banker's algorithm we should have pre-knowledge about three
things:-
- How
many instance of each resource a process could request. (Max)
- How
many instance of each resource is already allocated to that
process(Allocated)
- How
many instance of each resource is already available(Available).
We can calculate need of each process from above information:-
Need=Max - Allocated.
If
need<=available then request will be accepted otherwise it is denied and it
will check for next process in waiting queue.(See Life Cycle Of a Process)
Safe or
Unsafe State:- A system is in Safe state if its all process finish
its execution or if any process is unable to aquire its all requested
resources then system will be in Unsafe state.
Example:- To
understand banker's algorithm,we consider a system with four processes P1
through P4 and three resource type A,B and C.Resource type has 10 instances,B
has 5 instances and C has total 7 instances.
Current Scenario:-
Need[P2] < Available..So P2 we allocate resources to P2. After Completion of
P2 ,Scenario...
Now Need[P4] < Available so P4 will complete its execution
.After that--
So P1 will execute and after that P3 will take
resources.
Safe
Sequence is :- P2 , P4, P1,P3 (we
can obtain one more sequence if we allocate resources to P3 before P1 .)
Program:-
#include<stdio.h>
#include<conio.h>
int process,resource,i,j;
int avail[10],max[10][10],allot[10][10],need[10][10],completed[10];
void get_details()
{
int instanc;
//count,k variables are taken for counting purpose
printf("\n\t Enter No. of Process:-\n");
printf("\t\t");
scanf("%d",&process); //Entering No. of Processes
printf("\n\tEnter No. of Resources:-\n");
printf("\t\t");
scanf("%d",&resource); //No. of Resources
printf("\n\tEnter No. of Available Instances\n");
for(i=0;i<resource;i++)
{
printf("\t\t");
scanf("%d",&instanc);
avail[i]=instanc; // Storing Available instances
}
printf("\n\tEnter Maximum No. of instances of resources that a Process need:\n");
for(i=0;i<process;i++)
{
printf("\n\t For P[%d]",i);
for(j=0;j<resource;j++)
{
printf("\t");
scanf("%d",&instanc);
max[i][j]=instanc;
}
}
printf("\n\t Enter no. of instances already allocated to process of a resource:\n");
for(i=0;i<process;i++)
{
printf("\n\t For P[%d]\t",i);
for(j=0;j<resource;j++)
{
printf("\t\t");
scanf("%d",&instanc);
allot[i][j]=instanc;
need[i][j]=max[i][j]-allot[i][j]; //calculating Need of each process
}
}
}
int is_safe()
{
int count1=0,count2=0,k=0;
int avail2[10];
for(i=0;i<process;i++)
{ completed[i]=0;
} //Setting Flag for uncompleted Process
for(i=0;i<resource;i++)
{ avail2[i]=avail[i];
}
printf("\n\t Safe Sequence is:- \t");
while(count1!=process)
{
count2=count1;
for(i=0;i<process;i++)
{
for(j=0;j<resource;j++)
{
if(need[i][j]<=avail2[j])
{
k++;
}//if
}//for
if(k==resource && completed[i]==0 )
{
printf("P[%d]\t",i);
completed[i]=1;
for(j=0;j<resource;j++)
{
avail2[j]=avail2[j]+allot[i][j];
}//for
count1++;
}//if
k=0;
}//for
if(count1==count2)
{
printf("\t\t Stop ..After this.....Deadlock \n");
return 0;
//break;
}//if
}//while
return 1;
}
//if return value=1 then less than or equal
int lre(int first[],int second[])
{
int i=0;
for(i=0;i<resource;i++)
{
if(first[i]>second[i]){return 0;}
}
return 1;
}
void modify(int first[],int second[],char op)
{
int i=0;
for(i=0;i<resource;i++)
{
if(op=='+')
first[i]=first[i]+second[i];
else
first[i]=first[i]-second[i];
}
}
void print(int a[10][10])
{
for(i=0;i<process;i++)
{
for(j=0;j<resource;j++)
{
printf(" %d ",a[i][j]);
}
printf("\n");
}
}
void bankers()
{
int req[10],i=0,pid;
printf("\nenter requesting process number:");
scanf("%d",&pid);
printf("\nenter request vector:");
for(i=0;i<resource;i++)
{
scanf("%d",&req[i]);
}
if(lre(req,need[pid]))
{
if(lre(req,avail))
{
modify(avail,req,'-');
modify(allot[pid],req,'+');
modify(need[pid],req,'-');
printf("\nallocation:\n");
print(allot);
printf("\nneed:\n");
print(need);
if(is_safe())
{
printf("Requested Resources are allocated for P[%d]",pid);
}
else
{
modify(avail,req,'+');
modify(allot[pid],req,'-');
modify(need[pid],req,'+');
printf("DEADLOCK COMES->Requested Resources are NOT allocated for P[%d]",pid);
}
}
else
{
printf("\nP[%d] should wait -> no available resources",pid);
}
}
else
{
printf("\nsorry!!!error!!!process has requested more than its max claim");
}
}
void main()
{
clrscr();
get_details();
if(is_safe())
{
bankers();
}
getch();
}
#include<stdio.h>
#include<conio.h>
int process,resource,i,j;
int avail[10],max[10][10],allot[10][10],need[10][10],completed[10];
void get_details()
{
int instanc;
//count,k variables are taken for counting purpose
printf("\n\t Enter No. of Process:-\n");
printf("\t\t");
scanf("%d",&process); //Entering No. of Processes
printf("\n\tEnter No. of Resources:-\n");
printf("\t\t");
scanf("%d",&resource); //No. of Resources
printf("\n\tEnter No. of Available Instances\n");
for(i=0;i<resource;i++)
{
printf("\t\t");
scanf("%d",&instanc);
avail[i]=instanc; // Storing Available instances
}
printf("\n\tEnter Maximum No. of instances of resources that a Process need:\n");
for(i=0;i<process;i++)
{
printf("\n\t For P[%d]",i);
for(j=0;j<resource;j++)
{
printf("\t");
scanf("%d",&instanc);
max[i][j]=instanc;
}
}
printf("\n\t Enter no. of instances already allocated to process of a resource:\n");
for(i=0;i<process;i++)
{
printf("\n\t For P[%d]\t",i);
for(j=0;j<resource;j++)
{
printf("\t\t");
scanf("%d",&instanc);
allot[i][j]=instanc;
need[i][j]=max[i][j]-allot[i][j]; //calculating Need of each process
}
}
}
int is_safe()
{
int count1=0,count2=0,k=0;
int avail2[10];
for(i=0;i<process;i++)
{ completed[i]=0;
} //Setting Flag for uncompleted Process
for(i=0;i<resource;i++)
{ avail2[i]=avail[i];
}
printf("\n\t Safe Sequence is:- \t");
while(count1!=process)
{
count2=count1;
for(i=0;i<process;i++)
{
for(j=0;j<resource;j++)
{
if(need[i][j]<=avail2[j])
{
k++;
}//if
}//for
if(k==resource && completed[i]==0 )
{
printf("P[%d]\t",i);
completed[i]=1;
for(j=0;j<resource;j++)
{
avail2[j]=avail2[j]+allot[i][j];
}//for
count1++;
}//if
k=0;
}//for
if(count1==count2)
{
printf("\t\t Stop ..After this.....Deadlock \n");
return 0;
//break;
}//if
}//while
return 1;
}
//if return value=1 then less than or equal
int lre(int first[],int second[])
{
int i=0;
for(i=0;i<resource;i++)
{
if(first[i]>second[i]){return 0;}
}
return 1;
}
void modify(int first[],int second[],char op)
{
int i=0;
for(i=0;i<resource;i++)
{
if(op=='+')
first[i]=first[i]+second[i];
else
first[i]=first[i]-second[i];
}
}
void print(int a[10][10])
{
for(i=0;i<process;i++)
{
for(j=0;j<resource;j++)
{
printf(" %d ",a[i][j]);
}
printf("\n");
}
}
void bankers()
{
int req[10],i=0,pid;
printf("\nenter requesting process number:");
scanf("%d",&pid);
printf("\nenter request vector:");
for(i=0;i<resource;i++)
{
scanf("%d",&req[i]);
}
if(lre(req,need[pid]))
{
if(lre(req,avail))
{
modify(avail,req,'-');
modify(allot[pid],req,'+');
modify(need[pid],req,'-');
printf("\nallocation:\n");
print(allot);
printf("\nneed:\n");
print(need);
if(is_safe())
{
printf("Requested Resources are allocated for P[%d]",pid);
}
else
{
modify(avail,req,'+');
modify(allot[pid],req,'-');
modify(need[pid],req,'+');
printf("DEADLOCK COMES->Requested Resources are NOT allocated for P[%d]",pid);
}
}
else
{
printf("\nP[%d] should wait -> no available resources",pid);
}
}
else
{
printf("\nsorry!!!error!!!process has requested more than its max claim");
}
}
void main()
{
clrscr();
get_details();
if(is_safe())
{
bankers();
}
getch();
}
Wednesday, September 9, 2015
Atomic Transactions
·
System Model
·
Log-based
Recovery
·
Checkpoints
·
Concurrent
Atomic Transactions
System Model
·
Assures that
operations happen as a single logical unit of work, in its entirety, or not at
all
·
Related to
field of database systems
·
Challenge is
assuring atomicity despite computer
system failures
·
Transaction - collection of instructions or operations that performs single
logical function
·
Here we are concerned with changes to stable
storage – disk
·
Transaction is series of read and write operations
·
Terminated by commit (transaction successful) or abort (transaction
failed) operation
·
Aborted transaction must be rolled back to undo any
changes it performed
Types of Storage Media
·
Volatile
storage – information stored here does not survive system crashes
o Example: main memory, cache
·
Nonvolatile
storage – Information usually survives crashes
o Example: disk and tape
·
Stable storage
– Information never lost
o Not actually
possible, so approximated via replication or RAID to devices with independent failure modes
·
Goal is to assure transaction atomicity where failures cause loss
of information on volatile storage
Log-Based Recovery
·
Record to
stable storage information about all modifications by a transaction
·
Most common is write-ahead logging
·
Log on stable storage, each log record describes
single transaction write operation, including
o Transaction
name
o Data item name
o Old value
o New value
·
<Ti starts> written to log when transaction Ti starts
·
<Ti commits>
written when Ti commits
·
Log entry must reach stable storage before operation on data
occurs
Log-Based Recovery Algorithm
·
Using the log,
system can handle any volatile memory errors
o Undo(Ti) restores value of all data updated by Ti
o Redo(Ti) sets values of all data in transaction Ti to new values
·
Undo(Ti) and redo(Ti)
must be idempotent
o Multiple
executions must have the same result as one execution
·
If system
fails, restore state of all updated data via log
o
If log contains <Ti starts> without <Ti commits>, undo(Ti)
o
If log contains <Ti starts> and <Ti commits>, redo(Ti)
Checkpoints
·
Log could become long, and recovery could take
long->Checkpoints shorten log and recovery time.
·
Output a log
record <checkpoint> to the log on stable storage
·
Now recovery only
includes Ti, such that Ti started executing before the most recent checkpoint,
and all transactions after Ti All other transactions already on stable storage
Concurrent Transactions
·
Must be
equivalent to serial execution – serializability
·
Could perform all transactions in critical
section
·
Inefficient,
too restrictive
·
Concurrency-control
algorithms provide serializability
Serializability
·
Consider two
data items A and B
·
Consider
Transactions T0 and
T1
·
Execute T0, T1 atomically
·
Execution sequence
called schedule
·
Atomically
executed transaction order called serial schedule
·
For N
transactions, there are N! valid serial schedules
·
Schedule 1: T0
then T1
Nonserial Schedule
·
Nonserial schedule allows overlapped execute
·
Resulting
execution not necessarily incorrect
·
Consider
schedule S, operations Oi, Oj
·
Conflict if access same
data item, with at least one write
·
If Oi, Oj consecutive and operations of different
transactions & Oi and Oj don’t conflict
·
Then S’ with swapped order Oj Oi equivalent to S
·
If S can become
S’ via swapping nonconflicting operations.
·
S is conflict serializable
·
Schedule 2: Concurrent Serializable Schedule
Locking Protocol
·
Ensure serializability by associating lock with each data item
·
Follow locking
protocol for access control
·
Locks
o Shared – Ti has shared-mode lock (S) on item Q, Ti can read Q but not write Q
o Exclusive – Ti has
exclusive-mode lock (X) on Q, Ti can read and write Q
·
Require every
transaction on item Q acquire appropriate lock
·
If lock already
held, new request may have to wait
·
Similar to readers-writers algorithm
Two-phase Locking Protocol
·
Generally
ensures conflict serializability
·
Each transaction
issues lock and unlock requests in two phases
o Growing –
obtaining locks
o Shrinking –
releasing locks
·
Does not
prevent deadlock
Timestamp-based
Protocols
·
Select order
among transactions in advance – timestamp-ordering
·
Transaction Ti associated with timestamp TS(Ti) before Ti starts
o TS(Ti) < TS(Tj) if Ti entered system
before Tj
o TS can be
generated from system clock or as logical counter incremented at each entry of
transaction
·
Timestamps
determine serializability
order
o If TS(Ti) < TS(Tj), system must ensure
produced schedule equivalent to serial schedule where Ti appears before Tj
Timestamp-based Protocol
Implementation
·
Data item Q
gets two timestamps
o W-timestamp(Q)
– largest timestamp of any transaction that executed write(Q) successfully
o R-timestamp(Q)
– largest timestamp of successful read(Q)
o Updated
whenever read(Q) or write(Q) executed
·
Timestamp-ordering
protocol assures any conflicting read and write executed in
timestamp order
·
Suppose Ti
executes read(Q)
o If TS(Ti) < W-timestamp(Q), Ti
needs to read value of Q that was already overwritten
§ read operation
rejected and Ti rolled back
o If TS(Ti) ≥ W-timestamp(Q)
§ read executed,
R-timestamp(Q) set to max(R-timestamp(Q), TS(Ti))
Timestamp-ordering Protocol
·
Suppose Ti executes
write(Q)
o If TS(Ti) < R-timestamp(Q),
value Q produced by Ti was needed previously and Ti assumed it would never be produced
§ Write operation
rejected, Ti rolled back
o If TS(Ti) < W-tiimestamp(Q), Ti attempting to write obsolete value of Q
§ Write operation
rejected and Ti rolled back
o Otherwise, write executed
·
Any rolled back
transaction Ti is assigned new timestamp and restarted
·
Algorithm
ensures conflict serializability and freedom from deadlock
Schedule Possible Under Timestamp Protocol
Subscribe to:
Posts (Atom)