Bench run: plain / 31800922314

← plain home | dcompact →

generated (UTC): 2026-08-14T14:00:45.743291+00:00

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

Runner hardware/software

num=100000000 | key_size=8 | value_size=50

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. With ToplingDB dcompact, most of that CPU and memory cost can be offloaded.

/dev/shm usage (disk space; after db_bench + omit/scan)

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

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. RocksDB omit columns are derived-from-main (no separate omit pass).

RAM usage over time

Sampled once per second from /proc/statm plus open-file page cache. The bar below each plot shows stage name + duration (consecutive repeats merge, e.g. readseq×3); colored bands mark the same intervals. Stage names may overflow into a neighbor cell; a label under the bar is used only when names would overlap.

• 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 this process currently has open (cachestat(2) on /proc/pid/fd, deduped by inode; covers buffered read/write and mmap). Pages brought in only via buffered I/O are not charged to process RSS; mmap'd file pages may appear in both pagecache and RSS/shared, so the series can overlap. When reads hit this cache, they avoid disk I/O, and the DB benefits in its own way.
• 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.

Comparison: db_bench fillseq suite (perf)

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

Lazy load demo (scan; RocksDB v8.10 baseline)

zipkey* needs an extra omit pass: scan_omit_key/value enables lazy value load (no real value load). RocksDB has no omit API, so the baseline is readseq×3 already present in the main fill* suite (no extra pass). master omitted here (v8.10 is the stronger RocksDB baseline). Time ratio = v8.10 / zipkey*: >1 = zipkey* faster, <1 = zipkey* slower. nextwithkey is ToplingDB-only.

fillrandom-omit / scan

fillseq-omit / scan

memtablerep_bench: CSPPMemTable advantage

Focus: CSPP (ToplingDB) vs skiplist. Baseline = RocksDB v8.10 skiplist. Column "CSPP / v8.10": for throughput >1 = CSPP faster; for us/op and elapsed <1 = CSPP cheaper.

Per-engine details

ToplingDB zipkeyonly

db_bench (fillrandom+compact+readseq/readrandom)

db_bench (fillseq suite)

db_bench omit lazy-load (fillrandom DB)

db_bench omit lazy-load (fillseq DB)

memtablerep_bench (skiplist)

memtablerep_bench (cspp, ToplingDB only)

ToplingDB zipkeyvalue

db_bench (fillrandom+compact+readseq/readrandom)

db_bench (fillseq suite)

db_bench omit lazy-load (fillrandom DB)

db_bench omit lazy-load (fillseq DB)

RocksDB v8.10

db_bench (fillrandom+compact+readseq/readrandom)

db_bench (fillseq suite)

memtablerep_bench (skiplist)

RocksDB master (5fe22535)

db_bench (fillrandom+compact+readseq/readrandom)

db_bench (fillseq suite)

memtablerep_bench (skiplist)