Results directory: /opt/f16-performance/F16-ZFS-NVMe-Baseline โ 318 metric file(s)
Root: README.txt
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
F16 NVMe/ZFS Performance Baseline
Packaged: 2026-03-10T10:34:12-03:00
Hostname: zfs-klara
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
SUMMARY
โโโโโโโ
Mirror results : mirror_20260306_122723
RAIDZ2 results : raidz2_20260309_102455
Total benchmarks (JSON) : 636
Total metrics (NDJSON) : 318
DIRECTORY STRUCTURE
โโโโโโโโโโโโโโโโโโโ
F16-ZFS-NVMe-Baseline/
โโโ README.txt โ This file
โโโ mirror/
โ โโโ hardware-info/ โ Baseline snapshot (pre-benchmark)
โ โโโ software-info/ โ Baseline snapshot (pre-benchmark)
โ โโโ pool-config/ โ Pool creation commands + properties
โ โโโ fio-jobs/ โ All 53 .fio job files
โ โโโ benchmarks/{seq,rand,latency,mixed}/ โ fio JSON+ output
โ โโโ metrics/{seq,rand,latency,mixed}/ โ PCP NDJSON time-series
โโโ raidz2/
โโโ (same structure)
BENCHMARK MATRIX
โโโโโโโโโโโโโโโโ
A. Sequential Throughput : read/write ร QD{1,32} ร nj{1,4,8,16} ร 1M (16 jobs)
B. Random IOPS : randread/randwrite ร QD{1,4,16,32} ร nj{1,8,16,32} ร 4K (32 jobs)
C. Latency : randread/randwrite ร QD1 ร nj1 ร 4K (2 jobs)
D. Mixed 70/30 : randrw ร QD16 ร nj{1,4,8} ร 4K (3 jobs)
Total: 53 jobs ร 3 runs = 159 runs per layout
CONFIGURATION
โโโโโโโโโโโโโ
fio size : 64G
fio runtime : 180s + 10s ramp
ioengine : libaio, direct=1, time_based=1
ashift : 12
recordsize : 128K (dataset)
volblocksize: 8K (ZVOL)
compression : off
sync : standard
ARC : primarycache=all
FILES
โโโโโ
<layout>/benchmarks/<category>/*_runN.json โ fio JSON+ (IOPS, BW, latency)
<layout>/benchmarks/<category>/*_runN.txt โ fio human-readable summary
<layout>/benchmarks/<category>/*_runN.manifest.json โ run metadata + timestamps
<layout>/metrics/<category>/*_runN.ndjson โ PCP time-series (CPU, mem, disk, ZFS)
<layout>/pool-config/pool-design.txt โ Human-readable pool documentation
<layout>/fio-jobs/*.fio โ All fio job definitions
TOOLS
โโโโโ
To view metrics in a web dashboard:
python3 metrics_web.py -d F16-ZFS-NVMe-Baseline/ -p 8088
To generate summary tables:
python3 metrics_summary.py <ndjson_file>
To convert NDJSON to CSV/table:
python3 ndjson_to_table.py <ndjson_file>
Mode: packaged per-layout (mirror/ and raidz2/)