darshan-job-summary.pl.in 51.6 KB
Newer Older
1
#!/usr/bin/env perl
2
#
3
#  (C) 2015 by Argonne National Laboratory.
4 5 6
#      See COPYRIGHT in top-level directory.
#

7
# Set via configure
8
my $PREFIX="@prefix@";
9

10
use warnings;
11
use lib "@prefix@/lib";
12 13
use TeX::Encode;
use Encode;
14
use File::Temp qw/ tempdir /;
15
use File::Basename;
16
use Cwd;
17 18
use Getopt::Long;
use English;
19
use Number::Bytes::Human qw(format_bytes);
20
use POSIX qw(strftime);
21

22 23 24 25 26 27 28 29 30 31 32 33 34
#
# system commands used
#
my $darshan_parser = "$PREFIX/bin/darshan-parser";
my $pdflatex       = "pdflatex";
my $epstopdf       = "epstopdf";
my $cp             = "cp";
my $mv             = "mv";
my $gnuplot        = "gnuplot";

my $orig_dir = getcwd;
my $output_file = "summary.pdf";
my $verbose_flag = 0;
35
my $summary_flag = 0;
36 37 38 39 40
my $input_file = "";
my %posix_access_hash = ();
my %mpiio_access_hash = ();
my @access_size = ();
my %hash_files = ();
41
my $partial_flag = 0;
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

# data structures for calculating performance
my %hash_unique_file_time = ();
my $shared_file_time = 0;
my $total_job_bytes = 0;

process_args();

check_prereqs();

my $tmp_dir = tempdir( CLEANUP => !$verbose_flag );
if ($verbose_flag)
{
    print "verbose: $tmp_dir\n";
}

58
open(PARSE_OUT, "$darshan_parser --base --perf $input_file |") || die("Can't execute \"$darshan_parser $input_file\": $!\n");
59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

open(FA_READ, ">$tmp_dir/file-access-read.dat") || die("error opening output file: $!\n");
open(FA_WRITE, ">$tmp_dir/file-access-write.dat") || die("error opening output file: $!\n");
open(FA_READ_SH, ">$tmp_dir/file-access-read-sh.dat") || die("error opening output file: $!\n");
open(FA_WRITE_SH, ">$tmp_dir/file-access-write-sh.dat") || die("error opening output file: $!\n");

my $last_read_start = 0;
my $last_write_start = 0;

my $cumul_read_indep = 0;
my $cumul_read_bytes_indep = 0;

my $cumul_write_indep = 0;
my $cumul_write_bytes_indep = 0;

my $cumul_read_shared = 0;
my $cumul_read_bytes_shared = 0;

my $cumul_write_shared = 0;
my $cumul_write_bytes_shared = 0;

my $cumul_meta_shared = 0;
my $cumul_meta_indep = 0;

83
my $perf_layer = "";
84
my $perf_est = 0.0;
85
my $perf_mbytes = 0;
86 87 88
my $current_module = "";
my $stdio_perf_est = 0.0;
my $stdio_perf_mbytes = 0;
89

90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
my $first_data_line = 1;
my %file_record_hash = ();
my %fs_data = ();

while($line = <PARSE_OUT>)
{
    chomp($line);

    if ($line =~ /^\s*$/)
    {
        # ignore blank lines
    }
    elsif ($line =~ /^#/)
    {
        if ($line =~ /^# exe: /)
        {
106
            $f_save = "";
107
            ($junk, $cmdline) = split(':', $line, 2);
108

109 110
            # add escape characters if needed for special characters in
            # command line
111 112 113 114 115 116 117 118
            if ($cmdline =~ /<unknown args>/)
            {
                # fortran "<unknown args> seems to throw things off,
                # so we don't encode that if it's present
                $f_save = substr($cmdline, -14);
                $cmdline = substr($cmdline, 0, -14); 
            }
            $cmdline = encode('latex', $cmdline) . $f_save;
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
        }
        elsif ($line =~ /^# nprocs: /)
        {
            ($junk, $nprocs) = split(':', $line, 2);
        }
        elsif ($line =~ /^# run time: /)
        {
            ($junk, $runtime) = split(':', $line, 2);
        }
        elsif ($line =~ /^# start_time: /)
        {
            ($junk, $starttime) = split(':', $line, 2);
        }
        elsif ($line =~ /^# uid: /)
        {        
            ($junk, $uid) = split(':', $line, 2);
        }
        elsif ($line =~ /^# jobid: /)
        {
            ($junk, $jobid) = split(':', $line, 2);
        }
        elsif ($line =~ /^# darshan log version: /)
        {
            ($junk, $version) = split(':', $line, 2);
            $version =~ s/^\s+//;
        }
145 146
        elsif ($line =~ /^# agg_perf_by_slowest: /)
        {
147 148 149 150 151 152
            if($current_module eq "STDIO")
            {
                ($junk, $stdio_perf_est) = split(':', $line, 2);
                $stdio_perf_est = sprintf("%.2f", $stdio_perf_est);
            }
            elsif($current_module eq "POSIX")
153 154
            {
                ($junk, $perf_est) = split(':', $line, 2);
155
                $perf_est = sprintf("%.2f", $perf_est);
156 157
                $perf_layer = "POSIX";
            }
158
            elsif($current_module eq "MPI-IO")
159 160
            {
                ($junk, $perf_est) = split(':', $line, 2);
161
                $perf_est = sprintf("%.2f", $perf_est);
162 163 164
                $perf_layer = "MPI-IO";
            }
        }
165 166
        elsif ($line =~ /^# total_bytes: /)
        {
167 168 169 170 171 172 173
            if($current_module eq "STDIO")
            {
                ($junk, $perf_mbytes) = split(':', $line, 2);
                $stdio_perf_mbytes = $perf_mbytes / 1024 / 1024;
                $stdio_perf_mbytes = sprintf("%.1f", $stdio_perf_mbytes);
            }
            elsif($perf_mbytes == 0)
174 175 176 177 178 179
            {
                ($junk, $perf_mbytes) = split(':', $line, 2);
                $perf_mbytes = $perf_mbytes / 1024 / 1024;
                $perf_mbytes = sprintf("%.1f", $perf_mbytes);
            }
        }
180
        elsif ($line =~ /^# \*WARNING\*: .* contains incomplete data!/)
181 182 183
        {
            $partial_flag = 1;
        }
184 185 186 187
        elsif ($line =~ /^# (.+) module data/)
        {
            $current_module = $1;
        }
188 189 190 191 192 193 194
    }
    else
    {
        # parse line
        @fields = split(/[\t ]+/, $line);

        # encode the file system name to protect against special characters
195
        $fields[6] = encode('latex', $fields[6]);
196 197 198 199 200 201 202 203

        # is this our first piece of data?
        if($first_data_line)
        {
            $first_data_line = 0;
        }

        # is this a new file record?
204
        if (!defined $file_record_hash{$fields[2]})
205
        {
206 207 208
            $file_record_hash{$fields[2]} = ();
            $file_record_hash{$fields[2]}{FILE_NAME} = $fields[5];
            $file_record_hash{$fields[2]}{RANK} = $fields[1];
209 210
        }

211
        $file_record_hash{$fields[2]}{$fields[3]} = $fields[4];
212 213 214
        $summary{$fields[3]} += $fields[4];

        # accumulate independent and shared data as well as fs data
215
        if ($fields[3] eq "POSIX_F_READ_TIME" || $fields[3] eq "STDIO_F_READ_TIME")
216 217 218 219 220 221 222 223 224 225
        {
            if ($fields[1] == -1)
            {
                $cumul_read_shared += $fields[4];
            }
            else
            {
                $cumul_read_indep += $fields[4];
            }
        }
226
        elsif ($fields[3] eq "POSIX_F_WRITE_TIME" || $fields[3] eq "STDIO_F_WRITE_TIME")
227 228 229 230 231 232 233 234 235 236
        {
            if ($fields[1] == -1)
            {
                $cumul_write_shared += $fields[4];
            }
            else
            {
                $cumul_write_indep += $fields[4];
            }
        }
237
        elsif ($fields[3] eq "POSIX_F_META_TIME" || $fields[3] eq "STDIO_F_META_TIME")
238 239 240 241 242 243 244 245 246 247
        {
            if ($fields[1] == -1)
            {
                $cumul_meta_shared += $fields[4];
            }
            else
            {
                $cumul_meta_indep += $fields[4];
            }
        }
248
        elsif ($fields[3] eq "POSIX_BYTES_READ" || $fields[3] eq "STDIO_BYTES_READ")
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
        {
            if ($fields[1] == -1)
            {
                $cumul_read_bytes_shared += $fields[4];
            }
            else
            {
                $cumul_read_bytes_indep += $fields[4];
            }
            if (not defined $fs_data{$fields[6]})
            {
                $fs_data{$fields[6]} = [0,0];
            }
            $fs_data{$fields[6]}->[0] += $fields[4];
        }
264
        elsif ($fields[3] eq "POSIX_BYTES_WRITTEN" || $fields[3] eq "STDIO_BYTES_WRITTEN")
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
        {
            if ($fields[1] == -1)
            {
                $cumul_write_bytes_shared += $fields[4];
            }
            else
            {
                $cumul_write_bytes_indep += $fields[4];
            }
            if (not defined $fs_data{$fields[6]})
            {
                $fs_data{$fields[6]} = [0,0];
            }
            $fs_data{$fields[6]}->[1] += $fields[4];
        }

        # record start and end of reads and writes
282 283
        elsif ($fields[3] eq "POSIX_F_READ_START_TIMESTAMP" ||
            $fields[3] eq "STDIO_F_READ_START_TIMESTAMP")
284 285 286 287
        {
            # store until we find the end
            $last_read_start = $fields[4];
        }
288
        elsif (($fields[3] eq "POSIX_F_READ_END_TIMESTAMP" || $fields[3] eq "STDIO_F_READ_END_TIMESTAMP") && $fields[4] != 0)
289 290 291 292 293 294 295 296 297 298 299 300
        {
            # assume we got the read start already 
            my $xdelta = $fields[4] - $last_read_start;
            if($fields[1] == -1)
            {
                print FA_READ_SH "$last_read_start\t0\t$xdelta\t0\n";
            }
            else
            {
                print FA_READ "$last_read_start\t$fields[1]\t$xdelta\t0\n";
            }
        }
301 302
        elsif ($fields[3] eq "POSIX_F_WRITE_START_TIMESTAMP" ||
            $fields[3] eq "STDIO_F_WRITE_START_TIMESTAMP")
303 304 305 306
        {
            # store until we find the end
            $last_write_start = $fields[4];
        }
307
        elsif (($fields[3] eq "POSIX_F_WRITE_END_TIMESTAMP" || $fields[3] eq "STDIO_F_WRITE_END_TIMESTAMP") && $fields[4] != 0)
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385
        {
            # assume we got the write start already 
            my $xdelta = $fields[4] - $last_write_start;
            if($fields[1] == -1)
            {
                print FA_WRITE_SH "$last_write_start\t0\t$xdelta\t0\n";
            }
            else
            {
                print FA_WRITE "$last_write_start\t$fields[1]\t$xdelta\t0\n";
            }
        }

        # record common access counter info
        elsif ($fields[3] =~ /^POSIX_ACCESS(.)_ACCESS/)
        {
            $access_size[$1] = $fields[4];
        }
        elsif ($fields[3] =~ /^MPIIO_ACCESS(.)_ACCESS/)
        {
            $access_size[$1] = $fields[4];
        }
        elsif ($fields[3] =~ /^POSIX_ACCESS(.)_COUNT/)
        {
            my $tmp_access_size = $access_size[$1];
            if(defined $posix_access_hash{$tmp_access_size})
            {
                $posix_access_hash{$tmp_access_size} += $fields[4];
            }
            else
            {
                $posix_access_hash{$tmp_access_size} = $fields[4];
            }
        }
        elsif ($fields[3] =~ /^MPIIO_ACCESS(.)_COUNT/)
        {
            my $tmp_access_size = $access_size[$1];
            if(defined $mpiio_access_hash{$tmp_access_size})
            {
                $mpiio_access_hash{$tmp_access_size} += $fields[4];
            }
            else
            {
                $mpiio_access_hash{$tmp_access_size} = $fields[4];
            }
        }
    }
}
close(PARSE_OUT) || die "darshan-parser failure: $! $?";

# Fudge one point at the end to make xrange match in read and write plots.
# For some reason I can't get the xrange command to work.  -Phil
print FA_READ "$runtime\t-1\t0\t0\n";
print FA_WRITE "$runtime\t-1\t0\t0\n";
print FA_READ_SH "$runtime\t0\t0\t0\n";
print FA_WRITE_SH "$runtime\t0\t0\t0\n";
close(FA_READ);
close(FA_READ_SH);
close(FA_WRITE);
close(FA_WRITE_SH);

#
# Exit out if there are no actual file accesses
#
if ($first_data_line)
{
    $strtm = strftime("%a %b %e %H:%M:%S %Y", localtime($starttime));

    print "This darshan log has no file records. No summary was produced.\n";
    print "    jobid: $jobid\n";
    print "      uid: $uid\n";
    print "starttime: $strtm ($starttime )\n";
    print "  runtime: $runtime (seconds)\n";
    print "   nprocs: $nprocs\n";
    print "  version: $version\n";
    exit(1);
}

386
foreach $key (keys %file_record_hash)
387
{
388
    process_file_record($key, \%{$file_record_hash{$key}});
389 390 391 392 393 394 395 396 397 398
}

# copy template files to tmp tmp_dir
system "$cp $PREFIX/share/*.gplt $tmp_dir/";
system "$cp $PREFIX/share/*.tex $tmp_dir/";


# summary of time spent in POSIX & MPI-IO functions
open(TIME_SUMMARY, ">$tmp_dir/time-summary.dat") || die("error opening output file:$!\n");
print TIME_SUMMARY "# <type>, <app time>, <read>, <write>, <meta>\n";
399 400 401 402 403 404 405 406 407
if (defined $summary{POSIX_OPENS})
{
    print TIME_SUMMARY "POSIX, ", ((($runtime * $nprocs - $summary{POSIX_F_READ_TIME} -
        $summary{POSIX_F_WRITE_TIME} -
        $summary{POSIX_F_META_TIME})/($runtime * $nprocs)) * 100);
    print TIME_SUMMARY ", ", (($summary{POSIX_F_READ_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{POSIX_F_WRITE_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{POSIX_F_META_TIME}/($runtime * $nprocs))*100), "\n";
}
408 409 410 411 412 413 414 415 416
if (defined $summary{MPIIO_INDEP_OPENS})
{
    print TIME_SUMMARY "MPI-IO, ", ((($runtime * $nprocs - $summary{MPIIO_F_READ_TIME} -
        $summary{MPIIO_F_WRITE_TIME} -
        $summary{MPIIO_F_META_TIME})/($runtime * $nprocs)) * 100);
    print TIME_SUMMARY ", ", (($summary{MPIIO_F_READ_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{MPIIO_F_WRITE_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{MPIIO_F_META_TIME}/($runtime * $nprocs))*100), "\n";
}
417 418 419 420 421 422 423 424 425
if (defined $summary{STDIO_OPENS})
{
    print TIME_SUMMARY "STDIO, ", ((($runtime * $nprocs - $summary{STDIO_F_READ_TIME} -
        $summary{STDIO_F_WRITE_TIME} -
        $summary{STDIO_F_META_TIME})/($runtime * $nprocs)) * 100);
    print TIME_SUMMARY ", ", (($summary{STDIO_F_READ_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{STDIO_F_WRITE_TIME}/($runtime * $nprocs))*100);
    print TIME_SUMMARY ", ", (($summary{STDIO_F_META_TIME}/($runtime * $nprocs))*100), "\n";
}
426 427 428 429 430
close TIME_SUMMARY;

# counts of operations
open(PSX_OP_COUNTS, ">$tmp_dir/posix-op-counts.dat") || die("error opening output file: $!\n");
print PSX_OP_COUNTS "# <operation>, <POSIX count>\n";
Philip Carns's avatar
Philip Carns committed
431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
if (defined $summary{POSIX_OPENS})
{
    print PSX_OP_COUNTS
        "Read, ", $summary{POSIX_READS}, "\n",
        "Write, ", $summary{POSIX_WRITES}, "\n",
        "Open, ", $summary{POSIX_OPENS}, "\n",
        "Stat, ", $summary{POSIX_STATS}, "\n",
        "Seek, ", $summary{POSIX_SEEKS}, "\n",
        "Mmap, ", $summary{POSIX_MMAPS}, "\n",
        "Fsync, ", $summary{POSIX_FSYNCS} + $summary{POSIX_FDSYNCS}, "\n";
}
else
{
    print PSX_OP_COUNTS
        "Read, 0\n",
        "Write, 0\n",
        "Open, 0\n",
        "Stat, 0\n",
        "Seek, 0\n",
        "Mmap, 0\n",
        "Fsync, 0\n";
}
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
close PSX_OP_COUNTS;

if (defined $summary{MPIIO_INDEP_OPENS})
{
    # TODO: do we want to look at MPI split or non-blocking i/o here? 
    open(MPI_OP_COUNTS, ">$tmp_dir/mpiio-op-counts.dat") || die("error opening output file: $!\n");
    print MPI_OP_COUNTS "# <operation>, <MPI Ind. count>, <MPI Coll. count>\n";
    print MPI_OP_COUNTS
        "Read, ", $summary{MPIIO_INDEP_READS}, ", ", $summary{MPIIO_COLL_READS}, "\n",
        "Write, ", $summary{MPIIO_INDEP_WRITES}, ", ", $summary{MPIIO_COLL_WRITES}, "\n",
        "Open, ", $summary{MPIIO_INDEP_OPENS},", ", $summary{MPIIO_COLL_OPENS}, "\n",
        "Stat, ", "0, 0\n",
        "Seek, ", "0, 0\n",
        "Mmap, ", "0, 0\n",
        "Fsync, ", "0, ", $summary{MPIIO_SYNCS}, "\n";
    close MPI_OP_COUNTS;
}

Philip Carns's avatar
Philip Carns committed
471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
if (defined $summary{STDIO_OPENS})
{
    open(STDIO_OP_COUNTS, ">$tmp_dir/stdio-op-counts.dat") || die("error opening output file: $!\n");
    print STDIO_OP_COUNTS "# <operation>, <STDIO count>\n";
    print STDIO_OP_COUNTS
        "Read, ", $summary{STDIO_READS}, "\n",
        "Write, ", $summary{STDIO_WRITES}, "\n",
        "Open, ", $summary{STDIO_OPENS}, "\n",
        "Stat, 0\n",
        "Seek, ", $summary{STDIO_SEEKS}, "\n",
        "Mmap, 0\n",
        "Fsync, ", $summary{STDIO_FLUSHES}, "\n";
    close STDIO_OP_COUNTS;
}

486 487 488
# histograms of reads and writes (for POSIX and MPI-IO modules)
open (IO_HIST, ">$tmp_dir/posix-access-hist.dat") || die("error opening output file: $!\n");
print IO_HIST "# <size_range>, <POSIX_reads>, <POSIX_writes>\n";
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
if (defined $summary{POSIX_OPENS})
{
    print IO_HIST "0-100, ",
                  $summary{POSIX_SIZE_READ_0_100}, ", ",
                  $summary{POSIX_SIZE_WRITE_0_100}, "\n";
    print IO_HIST "101-1K, ",
                  $summary{POSIX_SIZE_READ_100_1K}, ", ",
                  $summary{POSIX_SIZE_WRITE_100_1K}, "\n";
    print IO_HIST "1K-10K, ",
                  $summary{POSIX_SIZE_READ_1K_10K}, ", ",
                  $summary{POSIX_SIZE_WRITE_1K_10K}, "\n";
    print IO_HIST "10K-100K, ",
                  $summary{POSIX_SIZE_READ_10K_100K}, ", ",
                  $summary{POSIX_SIZE_WRITE_10K_100K}, "\n";
    print IO_HIST "100K-1M, ",
                  $summary{POSIX_SIZE_READ_100K_1M}, ", ",
                  $summary{POSIX_SIZE_WRITE_100K_1M}, "\n";
    print IO_HIST "1M-4M, ",
                  $summary{POSIX_SIZE_READ_1M_4M}, ", ",
                  $summary{POSIX_SIZE_WRITE_1M_4M}, "\n";
    print IO_HIST "4M-10M, ",
                  $summary{POSIX_SIZE_READ_4M_10M}, ", ",
                  $summary{POSIX_SIZE_WRITE_4M_10M}, "\n";
    print IO_HIST "10M-100M, ",
                  $summary{POSIX_SIZE_READ_10M_100M}, ", ",
                  $summary{POSIX_SIZE_WRITE_10M_100M}, "\n";
    print IO_HIST "100M-1G, ",
                  $summary{POSIX_SIZE_READ_100M_1G}, ", ",
                  $summary{POSIX_SIZE_WRITE_100M_1G}, "\n";
    print IO_HIST "1G+, ",
                  $summary{POSIX_SIZE_READ_1G_PLUS}, ", ",
                  $summary{POSIX_SIZE_WRITE_1G_PLUS}, "\n";
}
else
{
    print IO_HIST "0-100, 0, 0\n";
    print IO_HIST "101-1K, 0, 0\n";
    print IO_HIST "1K-10K, 0, 0\n";
    print IO_HIST "10K-100K, 0, 0\n";
    print IO_HIST "100K-1M, 0, 0\n";
    print IO_HIST "1M-4M, 0, 0\n";
    print IO_HIST "4M-10M, 0, 0\n";
    print IO_HIST "10M-100M, 0, 0\n";
    print IO_HIST "100M-1G, 0, 0\n";
    print IO_HIST "1G+, 0, 0\n";
}
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573
close IO_HIST;

if (defined $summary{MPIIO_INDEP_OPENS})
{
    open (IO_HIST, ">$tmp_dir/mpiio-access-hist.dat") || die("error opening output file: $!\n");
    print IO_HIST "# <size_range>, <MPIIO_reads>, <MPIIO_writes>\n";
    print IO_HIST "0-100, ",
                  $summary{MPIIO_SIZE_READ_AGG_0_100}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_0_100}, "\n";
    print IO_HIST "101-1K, ",
                  $summary{MPIIO_SIZE_READ_AGG_100_1K}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_100_1K}, "\n";
    print IO_HIST "1K-10K, ",
                  $summary{MPIIO_SIZE_READ_AGG_1K_10K}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_1K_10K}, "\n";
    print IO_HIST "10K-100K, ",
                  $summary{MPIIO_SIZE_READ_AGG_10K_100K}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_10K_100K}, "\n";
    print IO_HIST "100K-1M, ",
                  $summary{MPIIO_SIZE_READ_AGG_100K_1M}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_100K_1M}, "\n";
    print IO_HIST "1M-4M, ",
                  $summary{MPIIO_SIZE_READ_AGG_1M_4M}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_1M_4M}, "\n";
    print IO_HIST "4M-10M, ",
                  $summary{MPIIO_SIZE_READ_AGG_4M_10M}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_4M_10M}, "\n";
    print IO_HIST "10M-100M, ",
                  $summary{MPIIO_SIZE_READ_AGG_10M_100M}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_10M_100M}, "\n";
    print IO_HIST "100M-1G, ",
                  $summary{MPIIO_SIZE_READ_AGG_100M_1G}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_100M_1G}, "\n";
    print IO_HIST "1G+, ",
                  $summary{MPIIO_SIZE_READ_AGG_1G_PLUS}, ", ",
                  $summary{MPIIO_SIZE_WRITE_AGG_1G_PLUS}, "\n";
    close IO_HIST;
}

574
    # sequential and consecutive access patterns
575 576
open (PATTERN, ">$tmp_dir/pattern.dat") || die("error opening output file: $!\n");
print PATTERN "# op total sequential consecutive\n";
577 578 579 580 581 582 583 584 585 586 587 588
if (defined $summary{POSIX_OPENS})
{
    print PATTERN "Read, ", $summary{POSIX_READS}, ", ",
        $summary{POSIX_SEQ_READS}, ", ", $summary{POSIX_CONSEC_READS}, "\n";
    print PATTERN "Write, ", $summary{POSIX_WRITES}, ", ",
        $summary{POSIX_SEQ_WRITES}, ", ", $summary{POSIX_CONSEC_WRITES}, "\n";
}
else
{
    print PATTERN "Read, 0, 0, 0\n";
    print PATTERN "Write, 0, 0, 0\n";
}
589 590 591 592 593
close PATTERN;

# table of common access sizes
open(ACCESS_TABLE, ">$tmp_dir/access-table.tex") || die("error opening output file:$!\n");
print ACCESS_TABLE "
594
\\begin{threeparttable}
595 596 597
\\begin{tabular}{r|r|r}
\\multicolumn{3}{c}{ } \\\\
\\multicolumn{3}{c}{Most Common Access Sizes} \\\\
598
\\multicolumn{3}{c}{(POSIX or MPI-IO)} \\\\
599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664
\\hline
\& access size \& count \\\\
\\hline
\\hline
";

# sort POSIX & MPI-IO access sizes (descending)
my $i = 0;
my $tmp_access_count = 0;
foreach $value (keys %posix_access_hash) {
    if ($posix_access_hash{$value} > 0) {
        $tmp_access_count++;
        if ($tmp_access_count == 4) {
            last;
        }
    }
}
if ($tmp_access_count > 0)
{
    foreach $value (sort {$posix_access_hash{$b} <=> $posix_access_hash{$a} } keys %posix_access_hash)
    {
        if ($i == 4) {
            last;
        }
        if ($posix_access_hash{$value} == 0) {
            last;
        }

        if ($i == 0) {
            print ACCESS_TABLE "
            \\multirow{$tmp_access_count}{*}{POSIX} \& $value \& $posix_access_hash{$value} \\\\\n
            ";
        }
        else {
            print ACCESS_TABLE "
            \& $value \& $posix_access_hash{$value} \\\\\n
            ";
        }
        $i++;
    }
}

$i = 0;
$tmp_access_count = 0;
foreach $value (keys %mpiio_access_hash) {
    if ($mpiio_access_hash{$value} > 0) {
        $tmp_access_count++;
        if ($tmp_access_count == 4) {
            last;
        }
    }
}
if ($tmp_access_count > 0)
{
    foreach $value (sort {$mpiio_access_hash{$b} <=> $mpiio_access_hash{$a} } keys %mpiio_access_hash)
    {
        if ($i == 4) {
            last;
        }
        if ($mpiio_access_hash{$value} == 0) {
            last;
        }

        if ($i == 0) {
            print ACCESS_TABLE "
            \\hline
665
            \\multirow{$tmp_access_count}{*}{MPI-IO \\textbf{\\ddag}} \& $value \& $mpiio_access_hash{$value} \\\\\n
666 667 668 669 670 671 672 673 674 675 676 677 678 679 680
            ";
        }
        else {
            print ACCESS_TABLE "
            \& $value \& $mpiio_access_hash{$value} \\\\\n
            ";
        }
        $i++;
    }
}

print ACCESS_TABLE "
\\hline
\\end{tabular}
";
681 682 683 684 685 686 687 688 689 690 691
if ($tmp_access_count > 0)
{
    print ACCESS_TABLE "
    \\begin{tablenotes}
    \\item[\\normalsize \\textbf{\\ddag}] \\scriptsize NOTE: MPI-IO accesses are given in terms of aggregate datatype size.
    \\end{tablenotes}
    ";
}
print ACCESS_TABLE "
\\end{threeparttable}
";
692 693 694
close ACCESS_TABLE;

# file count table
695 696 697 698 699
open(FILE_CNT_TABLE, ">$tmp_dir/file-count-table.tex") || die("error opening output file:$!\n");
print FILE_CNT_TABLE "
\\begin{tabular}{r|r|r|r}
\\multicolumn{4}{c}{ } \\\\
\\multicolumn{4}{c}{File Count Summary} \\\\
700
\\multicolumn{4}{c}{(estimated by POSIX I/O access offsets)} \\\\
701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
\\hline
type \& number of files \& avg. size \& max size \\\\
\\hline
\\hline
";

my $counter;
my $sum;
my $max;
my $key;
my $avg;

$counter = 0;
$sum = 0;
$max = 0;
foreach $key (keys %hash_files) {
    $counter++;
    if($hash_files{$key}{'min_open_size'} >
        $hash_files{$key}{'max_size'})
    {
        $sum += $hash_files{$key}{'min_open_size'};
        if($hash_files{$key}{'min_open_size'} > $max)
        {
            $max = $hash_files{$key}{'min_open_size'};
        }
    }
    else
    {
        $sum += $hash_files{$key}{'max_size'};
        if($hash_files{$key}{'max_size'} > $max)
        {
            $max = $hash_files{$key}{'max_size'};
        }
    }
}
if($counter > 0) { $avg = $sum / $counter; }
else { $avg = 0; }
$avg = format_bytes($avg);
$max = format_bytes($max);
print FILE_CNT_TABLE "total opened \& $counter \& $avg \& $max \\\\\n";

$counter = 0;
$sum = 0;
$max = 0;
foreach $key (keys %hash_files) {
    if($hash_files{$key}{'was_read'} && !($hash_files{$key}{'was_written'}))
    {
        $counter++;
        if($hash_files{$key}{'min_open_size'} >
            $hash_files{$key}{'max_size'})
        {
            $sum += $hash_files{$key}{'min_open_size'};
            if($hash_files{$key}{'min_open_size'} > $max)
            {
                $max = $hash_files{$key}{'min_open_size'};
            }
        }
        else
        {
            $sum += $hash_files{$key}{'max_size'};
            if($hash_files{$key}{'max_size'} > $max)
            {
                $max = $hash_files{$key}{'max_size'};
            }
        }
    }
}
if($counter > 0) { $avg = $sum / $counter; }
else { $avg = 0; }
$avg = format_bytes($avg);
$max = format_bytes($max);
print FILE_CNT_TABLE "read-only files \& $counter \& $avg \& $max \\\\\n";

$counter = 0;
$sum = 0;
$max = 0;
foreach $key (keys %hash_files) {
    if(!($hash_files{$key}{'was_read'}) && $hash_files{$key}{'was_written'})
    {
        $counter++;
        if($hash_files{$key}{'min_open_size'} >
            $hash_files{$key}{'max_size'})
        {
            $sum += $hash_files{$key}{'min_open_size'};
            if($hash_files{$key}{'min_open_size'} > $max)
            {
                $max = $hash_files{$key}{'min_open_size'};
            }
        }
        else
        {
            $sum += $hash_files{$key}{'max_size'};
            if($hash_files{$key}{'max_size'} > $max)
            {
                $max = $hash_files{$key}{'max_size'};
            }
        }
    }
}
if($counter > 0) { $avg = $sum / $counter; }
else { $avg = 0; }
$avg = format_bytes($avg);
$max = format_bytes($max);
print FILE_CNT_TABLE "write-only files \& $counter \& $avg \& $max \\\\\n";

$counter = 0;
$sum = 0;
$max = 0;
foreach $key (keys %hash_files) {
    if($hash_files{$key}{'was_read'} && $hash_files{$key}{'was_written'})
    {
        $counter++;
        if($hash_files{$key}{'min_open_size'} >
            $hash_files{$key}{'max_size'})
        {
            $sum += $hash_files{$key}{'min_open_size'};
            if($hash_files{$key}{'min_open_size'} > $max)
            {
                $max = $hash_files{$key}{'min_open_size'};
            }
        }
        else
        {
            $sum += $hash_files{$key}{'max_size'};
            if($hash_files{$key}{'max_size'} > $max)
            {
                $max = $hash_files{$key}{'max_size'};
            }
        }
    }
}
if($counter > 0) { $avg = $sum / $counter; }
else { $avg = 0; }
$avg = format_bytes($avg);
$max = format_bytes($max);
print FILE_CNT_TABLE "read/write files \& $counter \& $avg \& $max \\\\\n";

$counter = 0;
$sum = 0;
$max = 0;
foreach $key (keys %hash_files) {
    if($hash_files{$key}{'was_written'} &&
        $hash_files{$key}{'min_open_size'} == 0 &&
        $hash_files{$key}{'max_size'} > 0)
    {
        $counter++;
        if($hash_files{$key}{'min_open_size'} >
            $hash_files{$key}{'max_size'})
        {
            $sum += $hash_files{$key}{'min_open_size'};
            if($hash_files{$key}{'min_open_size'} > $max)
            {
                $max = $hash_files{$key}{'min_open_size'};
            }
        }
        else
        {
            $sum += $hash_files{$key}{'max_size'};
            if($hash_files{$key}{'max_size'} > $max)
            {
                $max = $hash_files{$key}{'max_size'};
            }
        }
    }
}
if($counter > 0) { $avg = $sum / $counter; }
else { $avg = 0; }
$avg = format_bytes($avg);
$max = format_bytes($max);
print FILE_CNT_TABLE "created files \& $counter \& $avg \& $max \\\\\n";

print FILE_CNT_TABLE "
\\hline
\\end{tabular}
";
close(FILE_CNT_TABLE);
877 878 879 880 881 882

# generate per filesystem data
open(FS_TABLE, ">$tmp_dir/fs-data-table.tex") || die("error opening output files:$!\n");
print FS_TABLE "
\\begin{tabular}{c|r|r|r|r}
\\multicolumn{5}{c}{ } \\\\
883
\\multicolumn{5}{c}{Data Transfer Per Filesystem (POSIX and STDIO)} \\\\
884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931
\\hline
\\multirow{2}{*}{File System} \& \\multicolumn{2}{c}{Write} \\vline \& \\multicolumn{2}{c}{Read} \\\\
\\cline{2-5}
\& MiB \& Ratio \& MiB \& Ratio \\\\\
\\hline
\\hline
";

foreach $key (keys %fs_data)
{
    my $wr_total_mb = ($fs_data{$key}->[1] / (1024*1024));
    my $rd_total_mb = ($fs_data{$key}->[0] / (1024*1024));

    my $wr_total_rt;
    if ($cumul_write_bytes_shared+$cumul_write_bytes_indep)
    {
        $wr_total_rt = ($fs_data{$key}->[1] / ($cumul_write_bytes_shared + $cumul_write_bytes_indep));
    }
    else
    {
        $wr_total_rt = 0;
    }

    my $rd_total_rt;
    if ($cumul_read_bytes_shared+$cumul_read_bytes_indep)
    {
        $rd_total_rt = ($fs_data{$key}->[0] / ($cumul_read_bytes_shared + $cumul_read_bytes_indep));
    }
    else
    {
        $rd_total_rt = 0;
    }

    printf FS_TABLE "%s \& %.5f \& %.5f \& %.5f \& %.5f \\\\\n",
        $key, $wr_total_mb, $wr_total_rt, $rd_total_mb, $rd_total_rt;
}

print FS_TABLE "
\\hline
\\end{tabular}
";
close FS_TABLE;

# variance data
open(VAR_TABLE, ">$tmp_dir/variance-table.tex") || die("error opening output file:$!\n");
print VAR_TABLE "
\\begin{tabular}{c|r|r|r|r|r|r|r|r|r}
\\multicolumn{10}{c}{} \\\\
932
\\multicolumn{10}{c}{Variance in Shared Files (POSIX and STDIO)} \\\\
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
\\hline
File \& Processes \& \\multicolumn{3}{c}{Fastest} \\vline \&
\\multicolumn{3}{c}{Slowest} \\vline \& \\multicolumn{2}{c}{\$\\sigma\$} \\\\
\\cline{3-10}
Suffix \&  \& Rank \& Time \& Bytes \& Rank \& Time \& Bytes \& Time \& Bytes \\\\
\\hline
\\hline
";

my $curcount = 1;
foreach $key (sort { $hash_files{$b}{'slowest_time'} <=> $hash_files{$a}{'slowest_time'} } keys %hash_files) {

    if ($curcount > 20) { last; }

    if ($hash_files{$key}{'procs'} > 1)
    {
        my $vt = sprintf("%.3g", sqrt($hash_files{$key}{'variance_time'}));
        my $vb = sprintf("%.3g", sqrt($hash_files{$key}{'variance_bytes'}));
        my $fast_bytes = format_bytes($hash_files{$key}{'fastest_bytes'});
        my $slow_bytes = format_bytes($hash_files{$key}{'slowest_bytes'});
953
        my $name = encode('latex', "..." . substr($hash_files{$key}{'name'}, -12));
954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990

        print VAR_TABLE "
               $name \&
               $hash_files{$key}{'procs'} \&
               $hash_files{$key}{'fastest_rank'} \&
               $hash_files{$key}{'fastest_time'} \&
               $fast_bytes \&
               $hash_files{$key}{'slowest_rank'} \&
               $hash_files{$key}{'slowest_time'} \&
               $slow_bytes \&
               $vt \&
               $vb \\\\
         ";
        $curcount++;
    }
}

print VAR_TABLE "
\\hline
\\end{tabular}
";
close VAR_TABLE;

# calculate performance
##########################################################################

# what was the slowest time by any proc for unique file access?
my $slowest_uniq_time = 0;
if(keys %hash_unique_file_time > 0)
{
    $slowest_uniq_time < $_ and $slowest_uniq_time = $_ for values %hash_unique_file_time;
}
print("Slowest unique file time: $slowest_uniq_time\n");
print("Slowest shared file time: $shared_file_time\n");
print("Total bytes read and written by app (may be incorrect): $total_job_bytes\n");
my $tmp_total_time = $slowest_uniq_time+$shared_file_time;
print("Total absolute I/O time: $tmp_total_time\n");
991
print("**NOTE: above shared and unique file times calculated using MPI-IO timers if MPI-IO interface used on a given file, POSIX timers otherwise.\n");
992

993 994 995 996 997 998
#Exit here if user ask only for a summary
if ($summary_flag)
{
        exit(0);
}

999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
# move to tmp_dir
chdir $tmp_dir;

# gather data to be used for document title (headers/footers)
($executable, $junk) = split(' ', $cmdline, 2);
@parts = split('/', $executable);
$cmd = $parts[$#parts];
@timearray = localtime($starttime);
$year = $timearray[5] + 1900;
$mon = $timearray[4] + 1;
$mday = $timearray[3];

# detect gnuplot ranges for file access graphs
my $ymax = $nprocs;
my $yinc = int($nprocs / 8);
if($yinc == 0) {$yinc=1;}
my $ymaxtic = $nprocs-1;

# reformat cumulative i/o data for file access table
my $cri = $cumul_read_indep / $nprocs;
my $crbi = $cumul_read_bytes_indep / ($nprocs * 1048576.0);

my $cwi = $cumul_write_indep / $nprocs;
my $cwbi = $cumul_write_bytes_indep / ($nprocs * 1048576.0);

my $crs = $cumul_read_shared / $nprocs;
my $crbs = $cumul_read_bytes_shared / ($nprocs * 1048576.0);

my $cws = $cumul_write_shared / $nprocs;
my $cwbs = $cumul_write_bytes_shared / ($nprocs * 1048576.0);

my $cmi = $cumul_meta_indep / $nprocs;
my $cms = $cumul_meta_shared / $nprocs;

# do any extra work needed for plotting mpi-io graphs
if (defined $summary{MPIIO_INDEP_OPENS})
{
1036
    system "$gnuplot -e \"data_file='mpiio-access-hist.dat'; graph_title='MPI-IO Access Sizes {/Times-Bold=32 \263}'; \\
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
    output_file='mpiio-access-hist.eps'\" access-hist-eps.gplt";
    system "$epstopdf mpiio-access-hist.eps";

    open(OP_COUNTS_PLT, ">>$tmp_dir/op-counts-eps.gplt") || die("error opening output file: $!\n");
    my $tmp_sz = -s "$tmp_dir/op-counts-eps.gplt";
    # overwrite existing newline
    truncate(OP_COUNTS_PLT, $tmp_sz-1);
    print OP_COUNTS_PLT ", \\
    \"mpiio-op-counts.dat\" using 2:xtic(1) title \"MPI-IO Indep.\", \\
    \"\" using 3 title \"MPI-IO Coll.\"\n";
    close OP_COUNTS_PLT;
}

Philip Carns's avatar
Philip Carns committed
1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061
# do any extra work needed for plotting stdio graphs
if (defined $summary{STDIO_OPENS})
{
    open(OP_COUNTS_PLT, ">>$tmp_dir/op-counts-eps.gplt") || die("error opening output file: $!\n");
    my $tmp_sz = -s "$tmp_dir/op-counts-eps.gplt";
    # overwrite existing newline
    truncate(OP_COUNTS_PLT, $tmp_sz-1);
    print OP_COUNTS_PLT ", \\
    \"stdio-op-counts.dat\" using 2:xtic(1) title \"STDIO\"\n";
    close OP_COUNTS_PLT;
}

1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
# execute base gnuplot scripts
system "$gnuplot time-summary-eps.gplt";
system "$epstopdf time-summary.eps";
system "$gnuplot op-counts-eps.gplt";
system "$epstopdf op-counts.eps";
system "$gnuplot -e \"data_file='posix-access-hist.dat'; graph_title='POSIX Access Sizes'; \\
output_file='posix-access-hist.eps'\" access-hist-eps.gplt";
system "$epstopdf posix-access-hist.eps";
system "$gnuplot -e \"ymax=$ymax; yinc=$yinc; ymaxtic=$ymaxtic; runtime='$runtime'\" file-access-eps.gplt";
system "$epstopdf file-access-read.eps";
system "$epstopdf file-access-write.eps";
system "$epstopdf file-access-shared.eps";
system "$gnuplot pattern-eps.gplt";
system "$epstopdf pattern.eps";

# generate summary PDF
# NOTE: we pass arguments to the latex template using '\def' commands
# NOTE: an autoconf test determines if -halt-on-error is available and sets
1080
# __DARSHAN_PDFLATEX_HALT_ON_ERROR accordingly
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099
my $latex_cmd_line = "\"\\def\\titlecmd{$cmd} \\
    \\def\\titlemon{$mon} \\
    \\def\\titlemday{$mday} \\
    \\def\\titleyear{$year} \\
    \\def\\titlecmdline{$cmdline} \\
    \\def\\jobid{$jobid} \\
    \\def\\jobuid{$uid} \\
    \\def\\jobnprocs{$nprocs} \\
    \\def\\jobruntime{$runtime} \\
    \\def\\filecri{$cri} \\
    \\def\\filecrbi{$crbi} \\
    \\def\\filecwi{$cwi} \\
    \\def\\filecwbi{$cwbi} \\
    \\def\\filecrs{$crs} \\
    \\def\\filecrbs{$crbs} \\
    \\def\\filecws{$cws} \\
    \\def\\filecwbs{$cwbs} \\
    \\def\\filecmi{$cmi} \\
    \\def\\filecms{$cms} \\
1100 1101
    \\def\\filecmi{$cmi} \\
    \\def\\perflayer{$perf_layer} \\
1102
    \\def\\perfest{$perf_est} \\
1103
    \\def\\perfbytes{$perf_mbytes} \\
1104 1105
    \\def\\stdioperfest{$stdio_perf_est} \\
    \\def\\stdioperfbytes{$stdio_perf_mbytes} \\
1106
    \\input{summary.tex}\" \\
1107
    @__DARSHAN_PDFLATEX_HALT_ON_ERROR@";
1108

1109 1110 1111 1112 1113 1114
if ($partial_flag == 1)
{
    my $partial_log_flags = "\\def\\incompletelog{1} \\";
    $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $partial_log_flags . substr($latex_cmd_line, 1);
}

1115 1116
if (defined $summary{MPIIO_INDEP_OPENS})
{
1117
    my $mpiio_latex_flags = "\\def\\inclmpiio{1} \\";
1118 1119 1120
    $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $mpiio_latex_flags . substr($latex_cmd_line, 1);
}

1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
if($perf_est > 0)
{
    my $perf_latex_flags = "\\def\\inclperf{1} \\";
    $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $perf_latex_flags . substr($latex_cmd_line, 1);

}

if($stdio_perf_est > 0)
{
    my $stdio_latex_flags = "\\def\\inclstdio{1} \\";
    $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $stdio_latex_flags . substr($latex_cmd_line, 1);

}
1134

1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
$system_rc = system "$pdflatex $latex_cmd_line > latex.output";
if($system_rc)
{
    print("LaTeX generation (phase1) failed [$system_rc], aborting summary creation.\n");
    print("error log:\n");
    system("tail latex.output");
    exit(1);
}
$system_rc = system "$pdflatex $latex_cmd_line > latex.output2";
if($system_rc)
{
    print("LaTeX generation (phase2) failed [$system_rc], aborting summary creation.\n");
    print("error log:\n");
    system("tail latex.output2");
    exit(1);
}

# get back out of tmp dir and grab results
chdir $orig_dir;
system "$mv $tmp_dir/summary.pdf $output_file";


sub process_file_record
{
1159 1160 1161
    my $hash = $_[0];
    my(%file_record) = %{$_[1]};
    my $rank = $file_record{'RANK'};
1162

1163
    if(!defined $file_record{'POSIX_OPENS'} && !defined $file_record{'STDIO_OPENS'})
1164 1165 1166 1167 1168
    {
        # ignore data records that don't have POSIX & MPI data
        return;
    }

1169 1170
    if((!defined $file_record{'POSIX_OPENS'} || $file_record{'POSIX_OPENS'} == 0) &&
        (!defined $file_record{'STDIO_OPENS'} || $file_record{'STDIO_OPENS'} == 0) &&
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182
        (!defined $file_record{'MPIIO_INDEP_OPENS'} ||
        ($file_record{'MPIIO_INDEP_OPENS'} == 0 && $file_record{'MPIIO_COLL_OPENS'} == 0)))
    {
        # file wasn't really opened, just stat probably
        return;
    }

    # record smallest open time size reported by any rank
    # XXX this isn't doable since dropping SIZE_AT_OPEN counter
    $hash_files{$hash}{'min_open_size'} = 0;

    # record largest size that the file reached at any rank
1183
    if(defined $file_record{'POSIX_OPENS'})
1184
    {
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198
        if(!defined($hash_files{$hash}{'max_size'}) ||
            $hash_files{$hash}{'max_size'} <  
            ($file_record{'POSIX_MAX_BYTE_READ'} + 1))
        {
            $hash_files{$hash}{'max_size'} = 
                $file_record{'POSIX_MAX_BYTE_READ'} + 1;
        }
        if(!defined($hash_files{$hash}{'max_size'}) ||
            $hash_files{$hash}{'max_size'} <  
            ($file_record{'POSIX_MAX_BYTE_WRITTEN'} + 1))
        {
            $hash_files{$hash}{'max_size'} = 
                $file_record{'POSIX_MAX_BYTE_WRITTEN'} + 1;
        }
1199
    }
1200 1201 1202

    # for stdio as well
    if(defined $file_record{'STDIO_OPENS'})
1203
    {
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217
        if(!defined($hash_files{$hash}{'max_size'}) ||
            $hash_files{$hash}{'max_size'} <  
            ($file_record{'STDIO_MAX_BYTE_READ'} + 1))
        {
            $hash_files{$hash}{'max_size'} = 
                $file_record{'STDIO_MAX_BYTE_READ'} + 1;
        }
        if(!defined($hash_files{$hash}{'max_size'}) ||
            $hash_files{$hash}{'max_size'} <  
            ($file_record{'STDIO_MAX_BYTE_WRITTEN'} + 1))
        {
            $hash_files{$hash}{'max_size'} = 
                $file_record{'STDIO_MAX_BYTE_WRITTEN'} + 1;
        }
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
    }

    # make sure there is an initial value for read and write flags
    if(!defined($hash_files{$hash}{'was_read'}))
    {
        $hash_files{$hash}{'was_read'} = 0;
    }
    if(!defined($hash_files{$hash}{'was_written'}))
    {
        $hash_files{$hash}{'was_written'} = 0;
    }

    if(defined $file_record{'MPIIO_INDEP_OPENS'} &&
        ($file_record{'MPIIO_INDEP_OPENS'} > 0 ||
        $file_record{'MPIIO_COLL_OPENS'} > 0))
    {
        # mpi file
        if($file_record{'MPIIO_INDEP_READS'} > 0 ||
            $file_record{'MPIIO_COLL_READS'} > 0 ||
            $file_record{'MPIIO_SPLIT_READS'} > 0 ||
            $file_record{'MPIIO_NB_READS'} > 0)
        {
            # data was read from the file
            $hash_files{$hash}{'was_read'} = 1;
        }
        if($file_record{'MPIIO_INDEP_WRITES'} > 0 ||
            $file_record{'MPIIO_COLL_WRITES'} > 0 ||
            $file_record{'MPIIO_SPLIT_WRITES'} > 0 ||
            $file_record{'MPIIO_NB_WRITES'} > 0)
        {
            # data was written to the file
            $hash_files{$hash}{'was_written'} = 1;
        }
    }
    else
    {
        # posix file
1255
        if((defined $file_record{'POSIX_READS'} && $file_record{'POSIX_READS'} > 0) || (defined $file_record{'STDIO_READS'} && $file_record{'STDIO_READS'} > 0))
1256 1257 1258 1259
        {
            # data was read from the file
            $hash_files{$hash}{'was_read'} = 1;
        }
1260
        if((defined $file_record{'POSIX_WRITES'} && $file_record{'POSIX_WRITES'} > 0) || (defined $file_record{'STDIO_WRITES'} && $file_record{'STDIO_WRITES'} > 0))
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270
        {
            # data was written to the file 
            $hash_files{$hash}{'was_written'} = 1;
        }
    }

    $hash_files{$hash}{'name'} = $file_record{FILE_NAME};

    if ($rank == -1)
    {
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294
        if(defined $file_record{'POSIX_OPENS'} && $file_record{'POSIX_OPENS'} > 0)
        {
            $hash_files{$hash}{'procs'}          = $nprocs;
            $hash_files{$hash}{'slowest_rank'}   = $file_record{'POSIX_SLOWEST_RANK'};
            $hash_files{$hash}{'slowest_time'}   = $file_record{'POSIX_F_SLOWEST_RANK_TIME'};
            $hash_files{$hash}{'slowest_bytes'}  = $file_record{'POSIX_SLOWEST_RANK_BYTES'};
            $hash_files{$hash}{'fastest_rank'}   = $file_record{'POSIX_FASTEST_RANK'};
            $hash_files{$hash}{'fastest_time'}   = $file_record{'POSIX_F_FASTEST_RANK_TIME'};
            $hash_files{$hash}{'fastest_bytes'}  = $file_record{'POSIX_FASTEST_RANK_BYTES'};
            $hash_files{$hash}{'variance_time'}  = $file_record{'POSIX_F_VARIANCE_RANK_TIME'};
            $hash_files{$hash}{'variance_bytes'} = $file_record{'POSIX_F_VARIANCE_RANK_BYTES'};
        }
        elsif(defined $file_record{'STDIO_OPENS'} && $file_record{'STDIO_OPENS'} > 0)
        {
            $hash_files{$hash}{'procs'}          = $nprocs;
            $hash_files{$hash}{'slowest_rank'}   = $file_record{'STDIO_SLOWEST_RANK'};
            $hash_files{$hash}{'slowest_time'}   = $file_record{'STDIO_F_SLOWEST_RANK_TIME'};
            $hash_files{$hash}{'slowest_bytes'}  = $file_record{'STDIO_SLOWEST_RANK_BYTES'};
            $hash_files{$hash}{'fastest_rank'}   = $file_record{'STDIO_FASTEST_RANK'};
            $hash_files{$hash}{'fastest_time'}   = $file_record{'STDIO_F_FASTEST_RANK_TIME'};
            $hash_files{$hash}{'fastest_bytes'}  = $file_record{'STDIO_FASTEST_RANK_BYTES'};
            $hash_files{$hash}{'variance_time'}  = $file_record{'STDIO_F_VARIANCE_RANK_TIME'};
            $hash_files{$hash}{'variance_bytes'} = $file_record{'STDIO_F_VARIANCE_RANK_BYTES'};
        }
1295 1296 1297
    }
    else
    {
1298 1299 1300 1301 1302 1303
        my $total_time = 0;
        my $total_bytes = 0;

        if(defined $file_record{'POSIX_OPENS'})
        {
            $total_time += $file_record{'POSIX_F_META_TIME'} +
1304 1305
                         $file_record{'POSIX_F_READ_TIME'} +
                         $file_record{'POSIX_F_WRITE_TIME'};
1306
            $total_bytes += $file_record{'POSIX_BYTES_READ'} +
1307
                          $file_record{'POSIX_BYTES_WRITTEN'};
1308 1309 1310 1311 1312 1313 1314 1315 1316
        }
        if(defined $file_record{'STDIO_OPENS'})
        {
            $total_time += $file_record{'STDIO_F_META_TIME'} +
                         $file_record{'STDIO_F_READ_TIME'} +
                         $file_record{'STDIO_F_WRITE_TIME'};
            $total_bytes += $file_record{'STDIO_BYTES_READ'} +
                          $file_record{'STDIO_BYTES_WRITTEN'};
        }
1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397

        if(!defined($hash_files{$hash}{'slowest_time'}) ||
           $hash_files{$hash}{'slowest_time'} < $total_time)
        {
            $hash_files{$hash}{'slowest_time'}  = $total_time;
            $hash_files{$hash}{'slowest_rank'}  = $rank;
            $hash_files{$hash}{'slowest_bytes'} = $total_bytes;
        }

        if(!defined($hash_files{$hash}{'fastest_time'}) ||
           $hash_files{$hash}{'fastest_time'} > $total_time)
        {
            $hash_files{$hash}{'fastest_time'}  = $total_time;
            $hash_files{$hash}{'fastest_rank'}  = $rank;
            $hash_files{$hash}{'fastest_bytes'} = $total_bytes;
        }

        if(!defined($hash_files{$hash}{'variance_time_S'}))
        {
            $hash_files{$hash}{'variance_time_S'} = 0;
            $hash_files{$hash}{'variance_time_T'} = $total_time;
            $hash_files{$hash}{'variance_time_n'} = 1;
            $hash_files{$hash}{'variance_bytes_S'} = 0;
            $hash_files{$hash}{'variance_bytes_T'} = $total_bytes;
            $hash_files{$hash}{'variance_bytes_n'} = 1;
            $hash_files{$hash}{'procs'} = 1;
            $hash_files{$hash}{'variance_time'} = 0;
            $hash_files{$hash}{'variance_bytes'} = 0;
        }
        else
        {
            my $n = $hash_files{$hash}{'variance_time_n'};
            my $m = 1;
            my $T = $hash_files{$hash}{'variance_time_T'};
            $hash_files{$hash}{'variance_time_S'} += ($m/($n*($n+$m)))*(($n/$m)*$total_time - $T)*(($n/$m)*$total_time - $T);
            $hash_files{$hash}{'variance_time_T'} += $total_time;
            $hash_files{$hash}{'variance_time_n'} += 1;

            $hash_files{$hash}{'variance_time'}    = $hash_files{$hash}{'variance_time_S'} / $hash_files{$hash}{'variance_time_n'};

            $n = $hash_files{$hash}{'variance_bytes_n'};
            $m = 1;
            $T = $hash_files{$hash}{'variance_bytes_T'};
            $hash_files{$hash}{'variance_bytes_S'} += ($m/($n*($n+$m)))*(($n/$m)*$total_bytes - $T)*(($n/$m)*$total_bytes - $T);
            $hash_files{$hash}{'variance_bytes_T'} += $total_bytes;
            $hash_files{$hash}{'variance_bytes_n'} += 1;

            $hash_files{$hash}{'variance_bytes'}    = $hash_files{$hash}{'variance_bytes_S'} / $hash_files{$hash}{'variance_bytes_n'};

            $hash_files{$hash}{'procs'} = $hash_files{$hash}{'variance_time_n'};
        }
    }

    # if this is a non-shared file, then add the time spent here to the
    # total for that particular rank
    # XXX mpiio or posix? should we do both or just pick mpiio over posix?
    if ($rank != -1)
    {
        # is it mpi-io or posix?
        if(defined $file_record{MPIIO_INDEP_OPENS} &&
            ($file_record{MPIIO_INDEP_OPENS} > 0 ||
            $file_record{MPIIO_COLL_OPENS} > 0))
        {
            # add up mpi times
            if(defined($hash_unique_file_time{$rank}))
            {
                $hash_unique_file_time{$rank} +=
                    $file_record{MPIIO_F_META_TIME} + 
                    $file_record{MPIIO_F_READ_TIME} + 
                    $file_record{MPIIO_F_WRITE_TIME};
            }
            else
            {
                $hash_unique_file_time{$rank} =
                    $file_record{MPIIO_F_META_TIME} + 
                    $file_record{MPIIO_F_READ_TIME} + 
                    $file_record{MPIIO_F_WRITE_TIME};
            }
        }
        else
        {
1398
            if(defined($file_record{POSIX_OPENS}))
1399
            {
1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414
                # add up posix times
                if(defined($hash_unique_file_time{$rank}))
                {
                    $hash_unique_file_time{$rank} +=
                        $file_record{POSIX_F_META_TIME} + 
                        $file_record{POSIX_F_READ_TIME} + 
                        $file_record{POSIX_F_WRITE_TIME};
                }
                else
                {
                    $hash_unique_file_time{$rank} =
                        $file_record{POSIX_F_META_TIME} + 
                        $file_record{POSIX_F_READ_TIME} + 
                        $file_record{POSIX_F_WRITE_TIME};
                }
1415
            }
1416
            if(defined($file_record{STDIO_OPENS}))
1417
            {
1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432
                # add up posix times
                if(defined($hash_unique_file_time{$rank}))
                {
                    $hash_unique_file_time{$rank} +=
                        $file_record{STDIO_F_META_TIME} + 
                        $file_record{STDIO_F_READ_TIME} + 
                        $file_record{STDIO_F_WRITE_TIME};
                }
                else
                {
                    $hash_unique_file_time{$rank} =
                        $file_record{STDIO_F_META_TIME} + 
                        $file_record{STDIO_F_READ_TIME} + 
                        $file_record{STDIO_F_WRITE_TIME};
                }
1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447
            }
        }
    }
    else
    {
        # cumulative time spent on shared files by slowest proc
        # is it mpi-io or posix?
        if(defined $file_record{MPIIO_INDEP_OPENS} &&
            ($file_record{MPIIO_INDEP_OPENS} > 0 ||
            $file_record{MPIIO_COLL_OPENS} > 0))
        {
            $shared_file_time += $file_record{'MPIIO_F_SLOWEST_RANK_TIME'};
        }
        else
        {
1448 1449 1450 1451 1452 1453 1454 1455
            if(defined $file_record{'POSIX_OPENS'} && $file_record{'POSIX_OPENS'} > 0)
            {
                $shared_file_time += $file_record{'POSIX_F_SLOWEST_RANK_TIME'};
            }
            elsif(defined $file_record{'STDIO_OPENS'} && $file_record{'STDIO_OPENS'} > 0)
            {
                $shared_file_time += $file_record{'STDIO_F_SLOWEST_RANK_TIME'};
            }
1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480
        }
    }

    my $mpi_did_read = 0;
    if (defined $file_record{MPIIO_INDEP_OPENS})
    {
        $mpi_did_read =
            $file_record{'MPIIO_INDEP_READS'} + 
            $file_record{'MPIIO_COLL_READS'} + 
            $file_record{'MPIIO_NB_READS'} + 
            $file_record{'MPIIO_SPLIT_READS'};
    }

    # add up how many bytes were transferred
    if(defined $file_record{MPIIO_INDEP_OPENS} &&
        ($file_record{MPIIO_INDEP_OPENS} > 0 ||
        $file_record{MPIIO_COLL_OPENS} > 0) && (!($mpi_did_read)))
    {
        # mpi file that was only written; disregard any read accesses that
        # may have been performed for sieving at the posix level
        $total_job_bytes += $file_record{'POSIX_BYTES_WRITTEN'}; 
    }
    else
    {
        # normal case
1481 1482 1483 1484 1485 1486 1487 1488 1489 1490
        if(defined $file_record{'POSIX_OPENS'})
        {
            $total_job_bytes += $file_record{'POSIX_BYTES_WRITTEN'} +
                $file_record{'POSIX_BYTES_READ'};
        }
        if(defined $file_record{'STDIO_OPENS'})
        {
            $total_job_bytes += $file_record{'STDIO_BYTES_WRITTEN'} +
            $file_record{'STDIO_BYTES_READ'};
        }
1491 1492
    }
}
1493

1494 1495
sub process_args
{
1496
    use vars qw( $opt_help $opt_output $opt_verbose $opt_summary);
1497 1498 1499

    Getopt::Long::Configure("no_ignore_case", "bundling");
    GetOptions( "help",
1500
        "output=s",
1501 1502
        "verbose",
        "summary");
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514

    if($opt_help)
    {
        print_help();
        exit(0);
    }

    if($opt_output)
    {
        $output_file = $opt_output;
    }

1515 1516 1517 1518 1519
    if($opt_verbose)
    {
        $verbose_flag = $opt_verbose;
    }

1520 1521 1522 1523 1524
    if($opt_summary)
    {
        $summary_flag = $opt_summary;
    }
    
1525 1526 1527 1528 1529 1530 1531
    # there should only be one remaining argument: the input file 
    if($#ARGV != 0)
    {
        print "Error: invalid arguments.\n";
        print_help();
        exit(1);
    }
1532

1533 1534
    $input_file = $ARGV[0];

1535 1536 1537 1538 1539
    # give default output file a similar name to the input file.
    #   log_name => log_name.pdf
    if (not $opt_output)
    {
        $output_file = basename($input_file);
1540
        $output_file .= ".pdf";
1541 1542
    }

1543 1544 1545
    return;
}

1546 1547 1548 1549 1550 1551
#
# Check for all support programs needed to generate the summary.
#
sub check_prereqs
{
    my $rc;
1552
    my $output;
1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564
    my @bins = ($darshan_parser, $pdflatex, $epstopdf,
                $gnuplot, $cp, $mv);
    foreach my $bin (@bins)
    {
        $rc = checkbin($bin);
        if ($rc)
        {
            print("error: $bin not found in PATH\n");
            exit(1);
        }
    }

1565
    # check gnuplot version
1566 1567 1568 1569 1570 1571
    $output = `$gnuplot --version`;
    if($? != 0)
    {
        print("error: failed to execute $gnuplot.\n");
        exit(1);
    }
1572

1573
    $output =~ /gnuplot (\d+)\.(\d+)/;
1574
    if($1 < 4 || ($1 < 5 && $2 < 2))
1575 1576 1577 1578 1579
    {
        print("error: detected $gnuplot version $1.$2, but darshan-job-summary requires at least 4.2.\n");
        exit(1);
    }

1580 1581 1582
    return;
}

1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617
#
# Execute which to see if the binary can be found in
# the users path.
#
sub checkbin($)
{
    my $binname = shift;
    my $rc;

    # save stdout/err
    open(SAVEOUT, ">&STDOUT");
    open(SAVEERR, ">&STDERR");

    # redirect stdout/error
    open(STDERR, '>/dev/null');
    open(STDOUT, '>/dev/null');
    $rc = system("which $binname");
    if ($rc)
    {
        $rc = 1;
    }
    close(STDOUT);
    close(STDERR);

    # suppress perl warning
    select(SAVEERR);
    select(SAVEOUT);

    # restore stdout/err
    open(STDOUT, ">&SAVEOUT");
    open(STDERR, ">&SAVEERR");

    return $rc;
}

1618 1619 1620 1621 1622 1623 1624 1625 1626
sub print_help
{
    print <<EOF;

Usage: $PROGRAM_NAME <options> input_file

    --help          Prints this help message
    --output        Specifies a file to write pdf output to
                    (defaults to ./summary.pdf)
1627
    --summary       Print a very succinct I/O timing summary and exit
1628 1629
    --verbose       Prints and retains tmpdir used for LaTeX output

1630 1631
Purpose:

1632 1633
    This script reads a Darshan output file generated by a job and
    generates a pdf file summarizing job behavior.
1634 1635 1636 1637

EOF
    return;
}