# API Maker WebSocket benchmarks : connections per server, memory per client, scaling > API Maker WebSocket benchmarks on 7 Linode servers : up to 3.66 million clients on one server, 8.5 to 8.8 KB per client, new clients and notifications a second, and how to scale. Source: https://docs.apimaker.dev/v4/docs/benchmarks/websockets/index.html WebSocket benchmarks · API Maker 4.0.0 · 7 servers and 3 clusters · measured 14 October 2026 **How many WebSocket users can API Maker hold ? As many as the memory has room for.** API Maker installed by its own installer on 7 Linode servers, from 1 to 8 vCPUs, its MongoDB and Redis on the same machine. A second server opened WebSocket clients, each with both tokens and a subscription of its own, until the server was full, and autocannon made the calls which notify them, 60 seconds a run : 50,167,783 clients let in, 214,875,630 notifications delivered. Every page shows the outputs and what each part of the server held in memory. Clients on one server · 8 vCPU, 32 GB 3,662,564 connected at once · 1 vCPU, 1 GB : 26,484 Memory per client 8.5 to 8.8KB API Maker, the kernel and Redis together New clients / second · 8 vCPU 10,939 both tokens and a subscription · 1 vCPU : 2,602 Notification delivered in 0.83ms p50, from the HTTP call to the client ## The answers - [How many connections can one server hold ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/index.html#one-server) : 26,484 on 1 vCPU and 1 GB, 3,662,564 on 8 vCPUs and 32 GB. Memory ends first, never the CPU. - [How much memory when a server is full ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/index.html#memory) : 8.5 to 8.8 KB per client in all. Every part of each server, with every client connected. - [How do we scale to N users ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/index.html#scale) : More servers behind a load balancer, one Redis they share. The picture, and the rule to count the servers. - [One big server or several small ones ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/index.html#big-or-small) : The same 8 vCPUs and 16 GB as one, two and four servers, measured : 2.02M, 1.92M, 1.62M clients. - [What does the architecture look like ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html#architecture) : The clients, a load balancer, API Maker servers which each hold their own clients, and the one Redis they share : the diagram, with what was measured. - [What happens when a server restarts ?](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/summary.html#restart) : 601,479 clients were all back after 189 s, in the memory they had before. When they all leave at once, their sockets are closed after 18.0 s. ## How many connections one server holds **WebSocket clients connected at once** · Shared CPU · Dedicated CPU [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html) · shared · $5 / month · 26,484 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html) · shared · $24 / month · 338,016 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html) · shared · $48 / month · 801,972 [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html) · shared · $192 / month · 3,662,564 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html) · dedicated · $36 / month · 338,006 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html) · dedicated · $72 / month · 795,888 [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html) · dedicated · $144 / month · 1,732,035 Each client connected with both tokens and one subscription of its own, MongoDB, Redis and two sandboxes on the same server. | Server | CPU · price / month | Clients connected | What ended first | API Maker | Whole server, used | Per client | Clients per $ a month | |---|---|--:|---|--:|--:|--:|--:| | [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html) | Shared · $5 | **26,484** | memory | 126 MB | 813 MB of 961 MB | 8.6 KB | 5,297 | | [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html) | Shared · $24 | **338,016** | memory | 1,110 MB | 3,417 MB of 3,916 MB | 8.5 KB | 14,084 | | [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html) | Shared · $48 | **801,972** | memory | 2,495 MB | 6,937 MB of 7,941 MB | 8.7 KB | 16,708 | | [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html) | Shared · $192 | **3,662,564** | memory | 10,196 MB | 27,618 MB of 32,096 MB | 8.8 KB | 19,076 | | [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html) | Dedicated · $36 | **338,006** | memory | 1,108 MB | 3,465 MB of 3,916 MB | 8.5 KB | 9,389 | | [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html) | Dedicated · $72 | **795,888** | memory | 2,474 MB | 6,917 MB of 7,941 MB | 8.7 KB | 11,054 | | [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html) | Dedicated · $144 | **1,732,035** | memory | 5,179 MB | 13,942 MB of 15,992 MB | 8.8 KB | 12,028 | A client which waits costs memory, and almost no CPU : the server pings it once a minute. So the number of clients a server holds follows its memory, shared or dedicated CPU alike. The page of each server gives every level of clients on the way, with what each part held. ## Memory when a server is full API Maker · MongoDB of API Maker · Redis · Sandboxes · Linux : its sockets, Docker, the rest · Available [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html#memory-at-full) 26k clients [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html#memory-at-full) 338k clients [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html#memory-at-full) 802k clients [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html#memory-at-full) 3.66M clients [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html#memory-at-full) 338k clients [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html#memory-at-full) 796k clients [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html#memory-at-full) 1.73M clients | Server | Clients | API Maker | Redis | Linux kernel | Per client : API Maker · kernel · Redis | Per client, in all | |---|--:|--:|--:|--:|--:|--:| | [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html#memory-at-full) | 26,484 | **126 MB** | 30 MB | 200 MB | 3.3 · 4.3 · 1.0 KB | **8.6 KB** | | [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html#memory-at-full) | 338,016 | **1,110 MB** | 356 MB | 1,635 MB | 3.1 · 4.3 · 1.0 KB | **8.5 KB** | | [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html#memory-at-full) | 801,972 | **2,495 MB** | 808 MB | 3,715 MB | 3.3 · 4.4 · 1.1 KB | **8.7 KB** | | [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html#memory-at-full) | 3,662,564 | **10,196 MB** | 3,533 MB | 16,443 MB | 3.2 · 4.5 · 1.1 KB | **8.8 KB** | | [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html#memory-at-full) | 338,006 | **1,108 MB** | 356 MB | 1,627 MB | 3.1 · 4.3 · 1.0 KB | **8.5 KB** | | [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html#memory-at-full) | 795,888 | **2,474 MB** | 802 MB | 3,691 MB | 3.3 · 4.4 · 1.1 KB | **8.7 KB** | | [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html#memory-at-full) | 1,732,035 | **5,179 MB** | 1,633 MB | 7,839 MB | 3.2 · 4.5 · 1.1 KB | **8.8 KB** | ## Scale to N users > Diagram : N WebSocket users : a load balancer, API Maker servers which each hold their clients, one Redis they share > > How API Maker serves N WebSocket users : the clients keep a WebSocket open to a load balancer, which passes each one to an API Maker server ; every server holds the clients which reached it, about 8 to 9 KB each ; all the servers share one Redis, which keeps a key and a channel per subscription and carries every notification to the server which holds the client, and one MongoDB. More users : more servers. A WebSocket stays on the server it reached, and what is said to a client can come from any server : the server which answers an API call, or runs `emitEventWS`, publishes the notification in Redis, and the server which holds the client sends it. So more users are more servers, and nothing else : no sticky session, no broker to add. How it was measured, and what ends first : [Scale to N users](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html). ## One big server or several small ones The same 8 dedicated vCPUs and 16 GB for API Maker, at the same price, as one server, two, and four ; MongoDB and Redis on a server of their own each time. | | [1 × 8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html#cluster-1x8) | [2 × 4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html#cluster-2x4) | [4 × 2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html#cluster-4x2) | |---|--:|--:|--:| | Clients connected at once | **2,022,026** | **1,915,960** | **1,619,372** | | … on each server | 2,022,026 | 957,980 | 404,843 | | Memory per client, in all | 8.8 KB | 8.8 KB | 8.6 KB | | New clients a second (both tokens, a subscription) | **12,412** | **12,696** | **10,297** | | … connect only, both tokens | 19,917 | 21,417 | 17,882 | | A notification for one person : delivery p50 | 1.01 ms | 1.18 ms | 1.37 ms | | … p99 | 1.72 ms | 1.80 ms | 2.51 ms | | Notifications a second, one person each | 24,305 | 24,977 | 25,566 | | Notifications a second, broadcast | 217,711 | 228,910 | 272,609 | | One server stops : clients dropped | 1,516,519 of 1,516,519 (100 %) | 383,192 of 766,384 (50 %) | 242,906 of 971,623 (25 %) | | … all back after | 221 s | 68.0 s | 38.0 s | Several servers hold about what one big server holds, and lose only their share of the clients when one of them stops. [The whole comparison](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html#big-or-small). ## New clients and notifications **New clients per second : both tokens and a subscription** · Shared CPU · Dedicated CPU [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html) · shared · $5 / month · 2,602 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html) · shared · $24 / month · 4,041 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html) · shared · $48 / month · 7,009 [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html) · shared · $192 / month · 10,939 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html) · dedicated · $36 / month · 3,464 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html) · dedicated · $72 / month · 6,627 [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html) · dedicated · $144 / month · 10,093 100 clients at a time connect, subscribe and leave, for 60 seconds. **Notifications per second : a broadcast to 1,000 clients** · Shared CPU · Dedicated CPU [1 vCPU · 1 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html) · shared · $5 / month · 72,245 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html) · shared · $24 / month · 103,387 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html) · shared · $48 / month · 147,974 [8 vCPU · 32 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html) · shared · $192 / month · 196,995 [2 vCPU · 4 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html) · dedicated · $36 / month · 69,874 [4 vCPU · 8 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html) · dedicated · $72 / month · 117,698 [8 vCPU · 16 GB](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html) · dedicated · $144 / month · 206,570 autocannon calls the emit event WS API as fast as the server takes ; what the clients received each second. ## The servers ### Shared CPU - [1 vCPU · 1 GB RAM · Shared CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-1-vcpu-1gb.html) : Linode Nanode 1 GB, $5 a month. 26,484 clients connected at once, 8.6 KB each. 2,602 new clients a second, a notification delivered in 2.51 ms. (14 October 2026 · 1,178,093 clients let in, 8,974,494 notifications) - [2 vCPU · 4 GB RAM · Shared CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-2-vcpu-4gb.html) : Linode 4 GB, $24 a month. 338,016 clients connected at once, 8.5 KB each. 4,041 new clients a second, a notification delivered in 1.45 ms. (14 October 2026 · 2,479,554 clients let in, 12,715,952 notifications) - [4 vCPU · 8 GB RAM · Shared CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-4-vcpu-8gb.html) : Linode 8 GB, $48 a month. 801,972 clients connected at once, 8.7 KB each. 7,009 new clients a second, a notification delivered in 1.06 ms. (14 October 2026 · 4,344,936 clients let in, 18,566,111 notifications) - [8 vCPU · 32 GB RAM · Shared CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-shared-8-vcpu-32gb.html) : Linode 32 GB, $192 a month. 3,662,564 clients connected at once, 8.8 KB each. 10,939 new clients a second, a notification delivered in 0.94 ms. (14 October 2026 · 8,562,774 clients let in, 22,866,139 notifications) ### Dedicated CPU - [2 vCPU · 4 GB RAM · Dedicated CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-2-vcpu-4gb.html) : Linode Dedicated 4 GB, $36 a month. 338,006 clients connected at once, 8.5 KB each. 3,464 new clients a second, a notification delivered in 1.74 ms. (14 October 2026 · 2,080,093 clients let in, 9,736,661 notifications) - [4 vCPU · 8 GB RAM · Dedicated CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-4-vcpu-8gb.html) : Linode Dedicated 8 GB, $72 a month. 795,888 clients connected at once, 8.7 KB each. 6,627 new clients a second, a notification delivered in 1.04 ms. (14 October 2026 · 4,102,260 clients let in, 15,486,511 notifications) - [8 vCPU · 16 GB RAM · Dedicated CPU](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/linode-dedicated-8-vcpu-16gb.html) : Linode Dedicated 16 GB, $144 a month. 1,732,035 clients connected at once, 8.8 KB each. 10,093 new clients a second, a notification delivered in 0.83 ms. (14 October 2026 · 6,038,773 clients let in, 23,621,794 notifications) ## How it is measured 1. **A real installation** API Maker is installed by its [installer](https://docs.apimaker.dev/v4/docs/getting-started/install-on-server.html) with its default answers, MongoDB and Redis in Docker on the same server. Nothing else runs : no reverse proxy, no other database. 2. **Real clients, from a second server** A dedicated server of the same data centre opens the WebSocket clients, from 11 addresses of its own (one address reaches one port of a server about 64,000 times). A client connects with both tokens, is told `CONNECTED`, sends one `REGISTER` with a condition of its own, and answers the pings of the server. 3. **autocannon makes the HTTP calls** The calls which notify the clients (the emit event WS API, the insert API of a table) and get all are made by autocannon, as in the API benchmarks. The clients count what they are told, and how late. 4. **60 seconds a run** Every measured run lasts 60 seconds, after a warm-up. Clients which were refused, notifications which did not arrive and calls which failed are counted. 5. **Clients are added until the server is full** Level after level, each held 30 seconds while the server is sampled (the last one 60). The run stops when what is available, less what Linux moved to the swap since the clients began to connect, is under 5 % of the memory of the server, or when it lets nobody else in : the clients fit in memory, not in the swap. 6. **Memory and CPU every second** The server samples its memory, its CPU, the API Maker process, every container, the sockets of the kernel and the table of its firewall while the clients are there. The whole method, the events, the can user connect codes and the commands : [How the WebSocket benchmarks are made](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/methodology.html). The data is the [API Maker benchmark dataset](https://apimaker.dev/json/benchmark_dataset.json) (`https://apimaker.dev/json/benchmark_dataset.json`). The APIs themselves : [API benchmarks](https://docs.apimaker.dev/v4/docs/benchmarks/index.html). ## Questions ### How many WebSocket connections can one API Maker server hold ? As many as its memory has room for : 26,484 on 1 shared vCPU and 1 GB, 338,016 on 2 shared vCPUs and 4 GB, 801,972 on 4 shared vCPUs and 8 GB, 3,662,564 on 8 shared vCPUs and 32 GB, 338,006 on 2 dedicated vCPUs and 4 GB, 795,888 on 4 dedicated vCPUs and 8 GB, 1,732,035 on 8 dedicated vCPUs and 16 GB, with MongoDB, Redis and two sandboxes on the same machine. A client which waits costs 8.5 to 8.8 KB of the server in all, and almost no CPU. ### How much memory does a WebSocket connection use ? 8.5 to 8.8 KB in all on the servers measured, for a client connected with both tokens and one subscription : most of it in the API Maker process, the rest in the Linux kernel (the socket) and in Redis (the subscription). ### How do I scale to N WebSocket users ? Add API Maker servers behind a load balancer : each holds the clients which reached it, and they share one Redis, which carries every notification to the server which holds the client. Count the servers from the memory : N users × 8.5 to 8.8 KB, plus one spare server. Measured : 1,915,960 clients on 2 servers, 1,619,372 on 4 servers. See [Scale to N users](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html). ### One big server or several small ones ? For the same price, they hold about the same number of clients : 2,022,026 on 1 × 8 vCPU · 16 GB, 1,915,960 on 2 × 4 vCPU · 8 GB, 1,619,372 on 4 × 2 vCPU · 4 GB. Several servers lose only a part of the clients when one of them stops, and can be restarted one at a time ; one big server has less to run. See [Scale to N users](https://docs.apimaker.dev/v4/docs/benchmarks/websockets/scaling.html). ### How fast are WebSocket notifications ? A notification for one person reached its client in 0.8 to 2.5 ms (p50) from the HTTP call which emitted it, at 1,000 notifications a second. In a broadcast, a server told up to 206,570 clients a second.