> For the complete documentation index, see [llms.txt](https://docs.gravity.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gravity.xyz/research-and-development/benchmark-results.md).
# 3-Validator Cluster Benchmark Results
3-validator cluster benchmark results from 2026-02-28, covering TPS benchmarks and bottleneck analysis across CPU, memory, and I/O.
### Test Environment
| Parameter | Value |
| ---------------- | ------------------------------- |
| **Topology** | 3 validators + 1 VFN |
| **Machine spec** | 8 vCPU / 16 GB RAM |
| **Disk** | 200 GB pd-balanced (4,200 IOPS) |
| **Accounts** | 100k accounts |
### Load Test Results (transaction removal rate)
* **Before write buffer is full**: \~**11k TPS** ERC20 transfer
* **After write buffer is full**: \~**9.5k TPS** ERC20 transfer
TPS over time
TPS detail
***
### Bottleneck Analysis
#### Resource Utilization
**1. CPU**
* CPU was **not** the bottleneck (usage \~**800%** on a 16-core machine)
* Execution, Merkle computation, and other pipeline stages all completed within the block time (\~**200 ms**)
CPU utilization during load test
**2. Memory**
* Memory grew continuously during the block cache fill phase, eventually reaching the **16 GB** limit
Memory growth as block cache fills
**3. Persistence (I/O)**
* Persistence latency was noticeably higher than the block time (\~**300 ms** vs \~**200 ms**)
* This caused the write buffer to accumulate and memory to grow
* Significant back-pressure appeared once the buffer reached the limit of **64**
Persistence latency vs. block time
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://docs.gravity.xyz/research-and-development/benchmark-results.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
