Performance Benchmarks

Benchmarking the Future of Timekeeping

ZEUS is designed to be deterministic first and fast second.

This page documents measured hashing performance for ZEUS Core using BLAKE3 under controlled local conditions.

---

Benchmark Environment

• Runtime: Node 18+

• Hashing primitive: BLAKE3

• Canonicalization: utf8_exact

• Test type: Sequential receipt generation

• Hardware: Modern consumer CPU (single process)

These numbers reflect local deterministic hashing performance, not network latency from the ZEUS Verify API.

---

Raw Hashing Results

1000

12 ms @ 0.012 ms per average operation

5000

10000

50000

Results demonstrate near-linear scaling characteristics under sequential execution.

---

What These Numbers Actually Mean

1

Linear Scaling Behavior

Performance scales approximately linearly with input volume.

There is no exponential slowdown under increasing load within expected operational ranges.

2

Hashing Is Not the Bottleneck

In production deployments, latency is typically dominated by:

• Network overhead

• Database writes

• Logging and audit storage

• TLS negotiation

The hashing operation itself remains computationally trivial relative to I/O.

3

Suitable for High-Frequency Use Cases

Given sub-millisecond per-operation cost, ZEUS can support:

• High-volume audit logging

• Event stream stamping

• Compliance trail construction

• Distributed system ordering

• Embedded device verification

Even constrained hardware environments can comfortably support thousands of receipts per second.

---

Important Clarifications

ZEUS does not claim:

• To outperform raw system clock reads

• To replace timekeeping hardware

• To eliminate network latency

ZEUS provides deterministic integrity over time data.

Speed is a property of the implementation. Integrity is the product.

---

API Considerations

When using ZEUS Verify over HTTPS:

• Expect additional latency from TLS and network routing.

• Hashing cost remains negligible compared to request roundtrip time.

• Throughput is governed by rate limits, not cryptographic cost.

API performance scales independently from local hashing performance.

---

On the 2038 Problem

ZEUS does not solve the 2038 problem.

ZEUS treats Unix time as input material, not as a foundational truth layer.

If Unix representations evolve, ZEUS receipts remain structurally verifiable because:

• The canonicalization format is explicit.

• The algorithm is labeled.

• The receipt version is embedded.

Temporal representation changes do not invalidate prior receipts.

---

Summary

ZEUS hashing performance is:

• Fast

• Linear

• Predictable

• Infrastructure-grade

However, performance is secondary.

The core value proposition is deterministic, cryptographic temporal structure.