c - Programatically Determining the File System Block Size -

this code i'm using time how time takes read file varying amount of bytes read.

for(int = 0; < trials; i++){              fd = open(argv[1], o_rdonly);             //set file offset random position on file             //but still multiple of current current test block size,             //simulating jumps of given test block size             //trying avoid prefetch             lseek(fd, test_block * rand() % (fs / test_block), seek_set);              //how time takes read `test_block` bytes             clock_gettime(clock_monotonic, &ts_ini);             ssize_t bytes_read = read(fd, buffer, test_block);             clock_gettime(clock_monotonic, &ts_end);              if(bytes_read > 0){                 accum += (((double)(ts_end.tv_sec - ts_ini.tv_sec)) +                      (ts_end.tv_nsec - ts_ini.tv_nsec)/nano_to_sec) / trials;             }             //closing file after each trial release resources             close(fd);         } 

now, if run program little shell script:

echo "block size(bytes) | avg. time(seconds)" block in 4 8 16 32 64 128 256 512 1024 2048 4096 8192 16384     ./bin/blocksize /tmp/random_gen $block done 

i result:

block size(bytes) | avg. time(seconds)                 4 | 0.002927567500                     8 | 0.003120735600                    16 | 0.004888980800                    32 | 0.003885210600                    64 | 0.003578379700                   128 | 0.001272970500                   256 | 0.004926633700                   512 | 0.001281894000                  1024 | 0.000243394200                  2048 | 0.000175361100                  4096 | 0.000001048200                  8192 | 0.000001938000                 16384 | 0.000003214000 

with result i'm assuming block size of system 4096 bytes (which agrees dumpe2fs), because @ point doesn't need additional operations, pass block got file, fast, , after times duplicates. (this guess)

but here weird part, if modify sh script little bit, adding clean caches before each execution, this:

echo "block size(bytes) | avg. time(seconds)" block in 4 8 16 32 64 128 256 512 1024 2048 4096 8192 16384     echo "echo 3 > /proc/sys/vm/drop_caches" | sudo sh     ./bin/blocksize /tmp/random_gen $block done 

then happens:

block size(bytes) | avg. time(seconds)                 4 | 0.006217417300                     8 | 0.003913319300                    16 | 0.004674101500                    32 | 0.005444699600                    64 | 0.005125086700                   128 | 0.004965967700                   256 | 0.002433360800                   512 | 0.002100266600                  1024 | 0.002221131400                  2048 | 0.001623008600                  4096 | 0.001936151500                  8192 | 0.001391976900                 16384 | 0.001270749800 

which doesn't make sense me. why times keeps decreasing test block size increases when clean cache first?

running on ubuntu 14.04lts 64bit

some points:

it possible filesystems have different block sizes on 1 system.

when rereading same file improved times because of caching. modern hd devices have onboard cache, , os has cache well.

posix provides standard way filesystem information block size: stavfs system call. stat, returns struct. shows 1 on system, each implementation may have extra/different fields, yours may differ:

 u_long      f_bsize;             /* preferred file system block size */  u_long      f_frsize;            /* fundamental filesystem block                                      (size if supported) */  fsblkcnt_t  f_blocks;            /* total # of blocks on file system                                      in units of f_frsize */  fsblkcnt_t  f_bfree;             /* total # of free blocks */  fsblkcnt_t  f_bavail;            /* # of free blocks avail                                      non-privileged user */  fsfilcnt_t  f_files;             /* total # of file nodes (inodes) */  fsfilcnt_t  f_ffree;             /* total # of free file nodes */  fsfilcnt_t  f_favail;            /* # of inodes avail                                      non-privileged user*/  u_long      f_fsid;              /* file system id (dev now) */  char        f_basetype[fstypsz]; /* target fs type name,                                      null-terminated */  u_long      f_flag;              /* bit mask of flags */  u_long      f_namemax;           /* maximum file name length */  char        f_fstr[32];          /* file system specific string */  u_long      f_filler[16];        /* reserved future expansion */ 

http://pubs.opengroup.org/onlinepubs/009695399/basedefs/sys/statvfs.h.html