Scribd
Upload
37 views2 pages

Elex - Case Study

The document discusses a case study on detecting deadlocks in distributed systems. It describes reports that estimate payroll expenses using multiple concurrent queries on an Employee table that are not always running. First, the database will be monitored to identify issues preventing the reports from running. Two stored procedures simultaneously update the Employee table, which could cause a deadlock. While deadlock detection algorithms have been proposed, issues around efficient resolution, correctness in distributed systems, and performance analysis need more research. Recommendations note that deadlock detection can proceed concurrently without impacting throughput but challenges include maintaining the wait-for-graph and detecting/resolving existing deadlocks.

Uploaded by

Vedanti Kesarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
37 views2 pages

Elex - Case Study

The document discusses a case study on detecting deadlocks in distributed systems. It describes reports that estimate payroll expenses using multiple concurrent queries on an Employee table that are not always running. First, the database will be monitored to identify issues preventing the reports from running. Two stored procedures simultaneously update the Employee table, which could cause a deadlock. While deadlock detection algorithms have been proposed, issues around efficient resolution, correctness in distributed systems, and performance analysis need more research. Recommendations note that deadlock detection can proceed concurrently without impacting throughput but challenges include maintaining the wait-for-graph and detecting/resolving existing deadlocks.

Uploaded by

Vedanti Kesarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Case Study on Deadlock detection in distributed systems.

Synopsis -
There is a database named HR1 that includes a table named Employee. The several read-only, historical reports
contain regularly changing totals. The reports use multiple queries to estimate payroll expenses. The queries run
concurrently. Users report that the payroll estimate reports do not always run.

Findings -
Firstly monitor the database to identify issues that prevent the reports from running.

Plan to deploy the application to a database server that supports other applications.

Must minimize the amount of storage that the database requires.

Employee Table -

The following Transact-SQL statements to create, configure, and populate the Employee table:

create table dbo.Employee

Here employee_id is the primary key. Following create an index and then insert it into the table using the SQL
command.

Discussion -
We have an application that updates the Employee's table. The application calls the following stored procedures
simultaneously and asynchronously:

✑ UspA: This stored procedure updates only the EmployeeStatus column.


✑ UspB: This stored procedure updates only the EmployeePayRate column.

The application uses views to control access to data. Views must meet the following requirements:

✑ Allow users access to all columns in the tables that the view accesses.

Restrict updates to only the rows that the view returns

Conclusion -
Although several deadlock detection algorithms have been proposed for distributed systems, a number of issues
remain to be addressed. future research should focus on the efficient resolution of deadlocks, the correctness of
distributed deadlock detection, modeling and performance analysis of deadlock detection algorithms, and the
probability of deadlocks in distributed systems.

Recommendations: SWOC
HELPFUL HARMFUL
Strengths: Weakness:
Once a cycle is formed in the state Deadlock is very sensitive to timing
graph, it persists until it is detected and requests. In DDBMS, message
broken. delays are unpredictable and there is
no global memory

SWOC
Opportunities: Challenges:
Once a cycle is formed in the state WFG maintenance and searching
graph, it persists until it is detected and the WFG for the presence of cycles.
broken. cycle detection can proceed Detecting existing deadlocks, and
concurrently with the normal activities second, resolving detected
of the system; therefore, it does not have deadlocks'
a have had serious effects on system
throughput.

You might also like