A program sorting a binary file of records in various ways using multiple processes.
The main goal of this program implementation was to practice with fork, exec, piping and sigaction.
Received statistics about turnaround time of processes and signals are printed at the end of the program execution.
Here's how the process tree looks like when making full use of the executables:
$ make clean
$ make
Command Line Parameters (and their flags) can be used in any order.
Command Line Flags h,q are optional and can be used up to 4 times. On the other hand, Command Line Flag f is required.
./mysort -f <Input File> -h|q <Column ID> [-h|q <Column ID>]
-
Each Record is represented by a simple struct of its data.
-
Coordinator node is represented by mysort executable. Coordinator node can fork and exec coach nodes according to the tree of processes shown above.
-
Coach nodes are represented by coach0, coach1, coach2 or coach3 executables. Coach nodes can fork and exec sorter nodes according to the tree of processes shown above.
-
Sorter nodes are represented by heapsorter or quicksorter executables.
-
Struct to define a record: records/record.c, records/record.h
-
Header file with important definitions: defines.h
-
Main function for Heapsort sorter: sorters/heapsorter.c
-
Main function for Quicksort sorter: sorters/quicksorter.c
-
Main function for coach with 2^0 = 1 sorter as child process: coaches/coach0.c
-
Main function for coach with 2^1 = 2 sorters as child processes: coaches/coach1.c
-
Main function for coach with 2^2 = 4 sorters as child processes: coaches/coach2.c
-
Main function for coach with 2^3 = 8 sorters as child processes: coaches/coach3.c
-
Main function for coordinator of 1 up to 4 coaches: coordinator.c
The records in output files are printed in the following form:
regNumber name surname houseStreet houseNumber city postcode salary