
==============================================================================
MERA BENCHMARK REPORT
==============================================================================
Date         : 2026-09-07
Host         : op4pgn
CPU          : INTEL(R) XEON(R) GOLD 6534 (32 threads)
RAM          : 1511.41 GB
Filesystem   : btrfs   mount /data1
Julia        : 1.12.7, 24 compute + 24 GC threads
Mera         : 1.8.0

Dataset      : L13_SN5_CD_only output 390
  ncpu       : 5120
  levelmax   : 13   (read capped at lmax=13)
  files      : 20489
  on disk    : 53.18 GB

Reading the RAMSES output (16 threads):
  hydro      :   447.95 s +- 6.97 s
  particles  :     3.61 s +- 1.31 s
  gravity    :    80.45 s +- 23.63 s
  TOTAL      :   532.00 s

Conversion (hydro, gravity, particles, 16 threads):
  read from RAMSES   :   514.86 s
  savedata write     :    27.57 s
  one-off conversion :   542.43 s
  MERA re-read warm  :  22.69 s   (22.7x faster)
  size on disk       : 53.18 GB -> 12.73 GB   (levels up to 13 only, not comparable)
  break-even         : 1.1 re-reads
  memory to load the same data:
    allocated RAMSES :  665.42 GB
    allocated MERA   :   47.42 GB   (14.0x less churn)
    GC RAMSES / MERA : 61.84 s / 1.63 s
    peak RSS RAMSES  :  115.17 GB
    peak RSS MERA    :   60.19 GB   (1.9x lower)

Storage by component:
  hydro        24.94 GB   46.9%
  gravity      17.52 GB   32.9%
  amr          10.66 GB   20.1%
  particle      65.0 MB    0.1%

Reading thread sweep (:hydro):
    1 threads :   709.61 s  1.00x
    2 threads :   556.15 s  1.28x
    4 threads :   483.96 s  1.47x
    8 threads :   464.68 s  1.53x
   16 threads :   442.08 s  1.61x
   24 threads :   460.02 s  1.54x
  fastest 16 threads, sweet spot 16 threads

Storage, IOPS by thread count:
    1 threads :     100336 IOPS
    2 threads :     174244 IOPS
    4 threads :     195868 IOPS
    8 threads :     237871 IOPS
   16 threads :     252068 IOPS
   24 threads :     283008 IOPS
==============================================================================
