Concurrency Control in Multilevel Relational Databases - Multilevel Security for Relational Databases - page 141

Table 7.1 Worklo ad Parameter

PARAMETER DESCRIPTION VALUE

Arrival rate The arrival rate of the transaction From 0 to 100

Clear levels Number of classification levels 2

Transaction size The average size of the transaction 10

captured at the machine using a monitoring tool provided by the
Microsoft SQL server.
An experimental database, the timesheet database, consisting of
four relations was created and populated to facilitate our performance
study. Timesheet system relations used in the implementation are
described as follows:

The employee relation provides information about employees:
Employees(EMPID, Code, Name, Department, Type,
Contract, Shift, Religion, Job, Position, Address, City)
The departure relation is used to store the departure notice of
each employee when he leaves the site of the work:
Departure(EmpID,

DepartureDate, ReturnDate,

D e p a r t u r eTy p e)
The timesheet relation is used to store the timesheet of each
employee every day:
TimeSheet(EMPID, Date, Timesheet, OverTime, Remarks)
The annual rights relation is used to store the rights of each
employee every year:
AnnualRights(EMPID, Year, Description, Inc, ADays, GDays)

Figure 7.3 shows an ER diagram for the timesheet system used
in the implementation of the prototype to facilitate our performance
study. The experiment investigates the average response times of trans-
actions at every security level for varying arrival rates. This experiment
uses the CPU response time (in milliseconds) as the metric.

7.4.3 Experiments and Results

In Figure 7.4, the response times of an enhanced secure multiver-
sion concurrency control model and the Rajwinder Singh model are
defined by using various arrival rates. In this figure, the response time
of the enhanced secure multiversion concurrency control model and