Bench run: dcompact / 31324418704

generated (UTC): 2026-08-09T17:28:01.168318+00:00

ToplingDB zipkeyonly db_bench | ToplingDB zipkeyvalue db_bench | RocksDB v8.10 db_bench | RocksDB master (5fe22535) db_bench | DB INFO LOGs (Actions artifact) — ToplingDB zipkeyonly LOG-fillrandom, ToplingDB zipkeyonly LOG-fillseq, ToplingDB zipkeyvalue LOG-fillrandom, ToplingDB zipkeyvalue LOG-fillseq, RocksDB v8.10 LOG-fillrandom, RocksDB v8.10 LOG-fillseq, RocksDB master (5fe22535) LOG-fillrandom, RocksDB master (5fe22535) LOG-fillseq

ToplingDB bench yaml (runtime graft): ToplingDB zipkeyonly db_bench-fillrandom.yaml | ToplingDB zipkeyonly db_bench-fillseq.yaml | ToplingDB zipkeyvalue db_bench-fillrandom.yaml | ToplingDB zipkeyvalue db_bench-fillseq.yaml

Bench configuration (dcompact)

Write-side db_bench CPU quota: 150% (from runner_env.txt cpu_quota_write). Topling compact: dcompact_worker — dcompact_worker runs outside the write cgroup and uses remaining CPU cores. RocksDB compact: compaction_service_spool — CompactionService spool + out-of-cgroup broker/worker on remaining cores.

Runner hardware/software

OS	Ubuntu 24.04.4 LTS
Kernel	6.17.0-1020-azure
CPU	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
Logical CPUs	4
Memory	15.6GiB
/dev/shm	7.8GiB
RocksDB block cache	7.8GiB

On-disk DB size (3.7GiB) ≤ block cache (7.8GiB) — cache can hold the entire dataset. So this benchmark mainly shows CPU and memory cost.

/dev/shm usage (disk space; after db_bench)

Allocated disk usage (IEC blocks). RocksDB uses default Snappy compression. Space ratio = engine / v8.10; <1 = less space than RocksDB, >1 = larger.

workload	ToplingDB zipkeyonly	ToplingDB zipkeyvalue	RocksDB v8.10	RocksDB master (5fe22535)	zipkeyonly / v8.10 (space)	zipkeyvalue / v8.10 (space)
fillrandom	3.0GiB	496.4MiB	3.7GiB	3.7GiB	0.81x	0.13x
fillseq	4.7GiB	1.1GiB	3.2GiB	3.2GiB	1.46x	0.35x

Peak RSS (RAM; during db_bench)

RSS is Resident Set Size. RocksDB block cache = half physical memory (7.8GiB). Ratio = engine / v8.10; <1 = less RSS, >1 = more RSS.

workload	ToplingDB zipkeyonly	ToplingDB zipkeyvalue	RocksDB v8.10	RocksDB master (5fe22535)	zipkeyonly / v8.10 (RSS)	zipkeyvalue / v8.10 (RSS)
fillrandom	956.3MiB	878.5MiB	4.1GiB	4.2GiB	0.23x	0.21x
fillseq	512.5MiB	1.2GiB	6.5GiB	6.6GiB	0.08x	0.19x

RAM usage over time

Sampled once per second from /proc/statm plus open-file page cache. Colored bands show benchmark segments (start time from db_bench output).

rss: resident set size (pages currently in RAM for the process); rss = shared + anony.
shared: shared resident pages; mostly readonly (cheap; OS prefers reclaiming these, no swap needed).
anony: rss - shared; mostly readwrite anonymous pages (costly, needs swap).
pagecache: kernel file page cache for regular files the process currently has open (cachestat(2) on /proc/pid/fd, deduped by inode; covers buffered readwrite and mmap). Pages brought in only via buffered I/O are not charged to process RSS; mmap'd file pages can appear in both pagecache and RSS/shared, so series may overlap.
anony+pc: anony+pagecache (sum of process anonymous RSS and open-file page cache; not a disjoint partition).

Comparison: db_bench fillrandom suite (perf)

Benchmarks: fillrandom, flush, compact, readseq×3, readrandom. RocksDB uses default Snappy compression. compact row shows operations/seconds. RocksDB time / zipkey*: >1 = that zipkey* engine faster. Values show ops/sec.

benchmark	ToplingDB zipkeyonly ops/sec	ToplingDB zipkeyvalue ops/sec	RocksDB v8.10 ops/sec	RocksDB master (5fe22535) ops/sec	zipkeyvalue time / zipkeyonly	v8.10 time / zipkeyonly	master (5fe22535) time / zipkeyonly	v8.10 time / zipkeyvalue	master (5fe22535) time / zipkeyvalue
compact	1/60.5s	1/75.0s	1/61.0s	1/45.5s	1.24x	1.01x	0.75x	0.81x	0.61x
fillrandom	672860	664376	496629	598401	1.01x	1.35x	1.12x	1.34x	1.11x
readrandom	4660558	1833600	376834	401778	2.54x	12.37x	11.60x	4.87x	4.56x
readseq	29045484	20495894	11026547	8523954	1.42x	1.67x	2.15x	1.18x	1.52x

Comparison: db_bench fillseq suite (perf)

Same as fillrandom. Watch compact / readseq / readrandom cost for minDictZip=10 vs space savings above.

benchmark	ToplingDB zipkeyonly ops/sec	ToplingDB zipkeyvalue ops/sec	RocksDB v8.10 ops/sec	RocksDB master (5fe22535) ops/sec	zipkeyvalue time / zipkeyonly	v8.10 time / zipkeyonly	master (5fe22535) time / zipkeyonly	v8.10 time / zipkeyvalue	master (5fe22535) time / zipkeyvalue
compact	1/87.7s	1/135.8s	1/35.3s	1/35.9s	1.55x	0.40x	0.41x	0.26x	0.26x
fillseq	1926705	1923782	3322359	3317752	1.00x	0.58x	0.58x	0.58x	0.58x
readrandom	6806775	1403633	361330	389766	4.85x	18.84x	17.46x	3.88x	3.60x
readseq	37876850	24577983	11314371	8574807	1.54x	3.35x	4.42x	2.17x	2.87x

Per-engine details

ToplingDB zipkeyonly

db_bench (fillrandom+compact+readseq/readrandom)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillrandom	1.486190	672860	148.619	100000000	37.2 MB/s
compact	60509343.000000	0	60.509	1
readseq	0.034464	29015986	3.446	100000000	1605.0 MB/s
readseq	0.034384	29081146	3.439	100000000	1608.6 MB/s
readseq	0.034427	29045484	3.443	100000000	1606.6 MB/s
readrandom	0.214565	4660558	21.457	100000000	163.0 MB/s (63212527 of 100000000 found)

db_bench (fillseq suite)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillseq	0.519020	1926705	51.902	100000000	106.6 MB/s
compact	87746996.000000	0	87.747	1
readseq	0.026519	37708954	2.652	100000000	2085.8 MB/s
readseq	0.026428	37837782	2.643	100000000	2092.9 MB/s
readseq	0.026399	37876850	2.640	100000000	2095.1 MB/s
readrandom	0.146912	6806775	14.691	100000000	376.5 MB/s (100000000 of 100000000 found)

ToplingDB zipkeyvalue

db_bench (fillrandom+compact+readseq/readrandom)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillrandom	1.505155	664376	150.517	100000000	36.7 MB/s
compact	75049781.000000	0	75.050	1
readseq	0.048922	20440241	4.892	100000000	1130.6 MB/s
readseq	0.048768	20504938	4.877	100000000	1134.2 MB/s
readseq	0.048790	20495894	4.879	100000000	1133.7 MB/s
readrandom	0.545373	1833600	54.538	100000000	64.1 MB/s (63207078 of 100000000 found)

db_bench (fillseq suite)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillseq	0.519809	1923782	51.981	100000000	106.4 MB/s
compact	135812979.000000	0	135.813	1
readseq	0.041454	24123042	4.145	100000000	1334.3 MB/s
readseq	0.040865	24470596	4.087	100000000	1353.5 MB/s
readseq	0.040687	24577983	4.069	100000000	1359.5 MB/s
readrandom	0.712436	1403633	71.244	100000000	77.6 MB/s (100000000 of 100000000 found)

RocksDB v8.10

db_bench (fillrandom+compact+readseq/readrandom)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillrandom	2.014	496629	201.358	100000000	27.5 MB/s
compact	60994529.000	0	60.995	1
readseq	0.131	7647461	8.266	63216431	423.0 MB/s
readseq	0.090	11089086	5.701	63216431	613.4 MB/s
readseq	0.091	11026547	5.733	63216431	609.9 MB/s
readrandom	2.654	376834	265.368	100000000	13.2 MB/s (63206659 of 100000000 found)

db_bench (fillseq suite)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillseq	0.301	3322359	30.099	100000000	183.8 MB/s
compact	35312178.000	0	35.312	1
readseq	0.147	6801494	14.703	100000000	376.2 MB/s
readseq	0.089	11295228	8.853	100000000	624.8 MB/s
readseq	0.088	11314371	8.838	100000000	625.8 MB/s
readrandom	2.768	361330	276.755	100000000	20.0 MB/s (100000000 of 100000000 found)

RocksDB master (5fe22535)

db_bench (fillrandom+compact+readseq/readrandom)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillrandom	1.671	598401	167.112	100000000	33.1 MB/s
compact	45545272.000	0	45.545	1
readseq	0.163	6122054	10.326	63216478	338.6 MB/s
readseq	0.118	8503588	7.434	63216478	470.4 MB/s
readseq	0.117	8523954	7.416	63216478	471.5 MB/s
readrandom	2.489	401778	248.893	100000000	14.0 MB/s (63217386 of 100000000 found)

db_bench (fillseq suite)

benchmark	micros/op	ops/sec	seconds	operations	extra
fillseq	0.301	3317752	30.141	100000000	183.5 MB/s
compact	35910497.000	0	35.911	1
readseq	0.183	5472260	18.274	100000000	302.7 MB/s
readseq	0.117	8567970	11.671	100000000	473.9 MB/s
readseq	0.117	8574807	11.662	100000000	474.3 MB/s
readrandom	2.566	389766	256.564	100000000	21.6 MB/s (100000000 of 100000000 found)