HttpArena
Comparisons

Kotlin web framework comparison: any two, head to head

ktor-ghost is quicker than ktor on 18 of the 29 profiles they both run, and behind on 11. On the HTTP/1.1 composite they score 2835 and 2590.

Pick any two of the 4 Kotlin entries. Requests per second is the best of three runs and the delta is the first column against the second; every row is the same profile at the same connection count, on the same machine. What each profile measures.

ProfileConnsktor-ghost req/secktor req/secDeltaktor-ghost p99ktor p99
Baseline4,096838,039822,062+2%9.60ms9.82ms
Pipelined4,096654,297584,867+12%146.90ms138.90ms
Short-lived4,096655,894665,691-1%54.10ms49.70ms
Async Delay32,000224,114244,271-8%225.40ms201.10ms
Latency-10K1,0249,9819,980+0%183.0us188.0us
Latency-1M1,024866,759878,822-1%2929664.0us2642944.0us
Latency-500K/81,024205,267197,383+4%11628544.0us11939840.0us
JSON Comp4,096217,405193,020+13%42.30ms82.30ms
JSON Comp16,384218,199189,511+15%145.20ms260.80ms
JSON TLS4,096607,943391,978+55%266.87ms11.92ms
8Gbit51249,24949,202+0%295.0us320.0us
Static TLS1,024191,799204,709-6%183.27ms245.95ms
Async DB1,024113,590103,951+9%11.20ms13.60ms
Fortunes1,02449,99151,030-2%43.10ms41.20ms
Baseline256515,664558,485-8%282.20ms44.86ms
Baseline1,024495,732541,980-9%768.20ms184.02ms
Static256196,780189,737+4%299.14ms41.76ms
Static1,024184,792180,206+3%530.61ms173.56ms
Baseline (h2c)256391,103412,245-5%359.42ms67.92ms
Baseline (h2c)1,024362,388402,964-10%1.95s210.39ms
Baseline (h2c)4,096392,445279,991+40%2.26s1.37s
JSON (h2c)1,024424,461314,284+35%319.32ms101.21ms
JSON (h2c)4,096389,919282,755+38%1.05s426.56ms
Echo5121,021,5631,021,279+0%1.33ms1.13ms
Echo4,096973,072980,019-1%9.46ms8.98ms
Echo16,384676,216645,400+5%42.90ms42.30ms
Echo Pipelined51211,253,40111,215,557+0%1.93ms2.13ms
Echo Pipelined4,09611,090,87110,647,634+4%14.40ms15.10ms
Echo Pipelined16,3848,286,8368,879,790-7%69.70ms61.10ms

Common comparisons

ktor-ghost vs ktor · ktor-ghost vs http4k · ktor-ghost vs neton · ktor vs http4k · ktor vs neton · http4k vs neton

Questions

Is ktor-ghost faster than ktor?

ktor-ghost is quicker than ktor on 18 of the 29 profiles they both run, and behind on 11. On the HTTP/1.1 composite they score 2835 and 2590. Both run on the same machine, in the same round, against the same profiles.

Is ktor-ghost faster than http4k?

ktor-ghost is quicker than http4k on 10 of the 11 profiles they both run, and behind on 1. On the HTTP/1.1 composite they score 2835 and 1901. Both run on the same machine, in the same round, against the same profiles.

Is ktor-ghost faster than neton?

ktor-ghost is quicker than neton on 8 of the 11 profiles they both run, and behind on 3. On the HTTP/1.1 composite they score 2835 and 1339. Both run on the same machine, in the same round, against the same profiles.

Is ktor faster than http4k?

ktor is quicker than http4k on 9 of the 11 profiles they both run, and behind on 1. On the HTTP/1.1 composite they score 2590 and 1901. Both run on the same machine, in the same round, against the same profiles.

Which Kotlin framework is fastest overall?

ktor-ghost holds the highest HTTP/1.1 composite of the Kotlin entries, at 2835. The composite is throughput over the whole suite; if a service is one shape of workload, compare that profile on its own in the table above, because the order changes.

Every Kotlin entry

Results updated · Kotlin composite ranking · Compare across languages · Other languages · Open the leaderboard