Brief example showing the read performance difference between random vs linear disk reads

disk_benchmark.c

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mman.h>
#include <unistd.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/types.h>

const unsigned int num_records = 100000000;

struct inventory
{
    unsigned int seq;
    unsigned int amount;
};

void create_data()
{
    struct inventory record;
    FILE *fd;

    fd = fopen("test.bin", "wb");

    for (unsigned int i = 0; i < num_records; i++)
    {
        memset(&record, 0, sizeof(struct inventory));
        record.seq = i;
        record.amount = rand() % 3;
        fwrite(&record, sizeof(struct inventory), 1, fd);
    }

    fclose(fd);
}

void mmap_read()
{
    size_t pagesize = getpagesize();
    printf("System page size: %zu bytes\n", pagesize);

    int fdin;
    char *src, *dst;
    struct stat statbuf;
    fdin = open("test.bin", O_RDONLY);
    fstat(fdin, &statbuf);
    src = mmap(0, statbuf.st_size, PROT_READ, MAP_SHARED, fdin, 0);

    struct inventory record;

    unsigned int recordno;
    for (unsigned int i = 0; i < 100000; i++)
    {

        // Use this to select random records from the file
        recordno = rand() % num_records;

        // Use this to read data linerally from the file
        // recordno = i;
        memcpy(&record, src + (sizeof(struct inventory) * recordno), sizeof(struct inventory));
        printf("%d %d\n", record.seq, record.amount);
    }
}

int main(int argc, char **argv)
{
    // Make sure to run `sync && sudo purge` between runs to flush the OS disk cache
    // https://stackoverflow.com/questions/28845524/echo-3-proc-sys-vm-drop-caches-on-mac-osx

    // Run this first to create the example data
    // create_data();

    // Run this to read 100k items from that file
    mmap_read();

    printf("hello world\n");
    return 0;
}