April 16, 2026

WireGuard Benchmark on MikroTik: Real-World Tests on 10 Devices with OptiWize SDWAN

WireGuard is faster! … but by how much?

AI-translated from Italian. Have a correction? Drop us a line.

WireGuard is faster! … but by how much?

We put WireGuard to the test on 10 MikroTik devices with OptiWize SDWAN. Here are the real results, numbers in hand.

WireGuard is faster! … but how much faster? Over the past few months, at OptiWize we've indulged ourselves in answering a question we've been carrying around for a while: "How fast does WireGuard actually run on MikroTik?" On paper it's lightweight, modern, and kernel-integrated. But in the real world — the world of ISPs, WISPs, and system integrators — numbers matter. And they need to be reliable. So we did what we do best: systematic benchmarking, comparative analysis across 10 devices, deep tuning … and a lot of technical sweat. Because spinning up a VPN tunnel is one thing; pushing it to its absolute limits without compromising stability and observability is an entirely different story.

Objectives of the Tests

  • Verify the maximum achievable throughput on each device with WireGuard configuration optimized via OptiWize SDWAN.
  • Simulate real-world traffic scenarios, both toward zero-latency targets and external targets (via Internet).
  • Understand how far you can push things and which models are worth investing in today.

The Setup

  • 10 MikroTik device models, from the entry-level range (RB3011, L009) up to top-tier models (CCR2004).
  • Tests performed with OptiWize SDWAN, in ECMP, multi-gateway configuration with native supervision.
  • All tests executed with CPU monitored in real-time and saturated at 100%, to ensure a realistic maximum figure.
  • Two scenarios:
    • Zero-latency target: throughput under optimal conditions (without traversing the Internet).
    • External target (WAN) at 12ms: throughput under real-world conditions, with full routing.
DeviceTX Max (Mbps)RX Max (Mbps)Active Gateways
CCR20041385.91518.23–4
RB4011-SFPp-19001298.51351.24
RB5009-251142.51148.42–4
RB5009-SFPp1122.61134.34
RB4011-SFPp1022.5997.74
HAPAX3-25981.01112.04
CCR1009370.1388.82–3
RB3011142.6217.43
L009136.3196.01–2

Nearly all "modern" devices comfortably exceed 1 Gbit/s, with CPU fully saturated. The limits are determined by the hardware and how the protocol has been implemented on that hardware, not by the protocol itself. WireGuard is being fully leveraged.

Results (External Target – Internet) Performance drops slightly here due to Internet traversal, but remains very solid:

  • RB4011-SFPp-1900 → ~977 Mbps TX / ~935 Mbps RX
  • CCR2004 → ~963 Mbps TX / ~804 Mbps RX
  • RB5009-SFPp → ~944 Mbps TX / ~599 Mbps RX

Important note: We identified a key effect worth highlighting: the higher the latency, the lower the WireGuard tunnel performance. This is expected behavior, but is often underestimated. It is therefore strategic to leverage one of the core capabilities of OptiWize SDWAN: the ability to choose the POP closest to the customer's Internet access point. This allows you to minimize latency and therefore maximize real perceived throughput, especially under simultaneous upload/download or VoIP/video traffic.

What's Behind the Numbers These figures are not "out of the box." They are the result of:

  • Kernel-level optimization of configurations.
  • Multi-gateway load balancing with dynamic failover logic.
  • Resource control scripts to prevent bottlenecks.
  • Extensive use of OptiWize SDWAN features, now running in production across dozens of installations.

In practice: all the know-how built up within our team, engineered to become automated provisioning.

Conclusions WireGuard, properly configured and orchestrated, is not a bottleneck. It is a key resource for ensuring security, performance, and simplicity in managing complex networks. With OptiWize SDWAN, even small MSPs or ISPs can:

  • Create secure VPNs in just a few clicks
  • Perform transparent failover or link aggregation
  • Keep everything monitored and centralized
  • Guarantee end customers a fast, resilient, and secure experience

If you want to know how to use these numbers to optimize your network or your customers' networks, get in touch. This isn't theory: everything is already ready in the platform.

💡 Want a demo? Write to us. Want to try it on your own device? Let's talk.

Device

TX Max (Mbps)

RX Max (Mbps)

Active Gateways

CCR2004

1385.9

1518.2

3–4

RB4011-SFPp-1900

1298.5

1351.2

4

RB5009-25

1142.5

1148.4

2–4

RB5009-SFPp

1122.6

1134.3

4

RB4011-SFPp

1022.5

997.7

4

HAPAX3-25

981.0

1112.0

4

CCR1009

370.1

388.8

2–3

RB3011

142.6

217.4

3

L009

136.3

196.0

1–2

Subscribe to our newsletter!

Stay up to date on industry topics and OptiWize features

    • Required field
  • Enter a valid email address

  • Enter a valid name

  • Enter a valid phone number

  • Value is too small

    • Required field
  • Enter a valid email address

  • Enter a valid name

  • Enter a valid phone number

  • Value is too small

By clicking the button you agree to our Privacy Policy

Want to explore a topic in more depth? Fill out the form and we'll get back to you as soon as possible

Email First Name Last Name Mobile Number of devices

0

10000+

    • Required field
  • Enter a valid email address

  • Enter a valid name

  • Enter a valid phone number

  • Value is too small

    • Required field
  • Enter a valid email address

  • Enter a valid name

  • Enter a valid phone number

  • Value is too small

WireGuard Benchmark on MikroTik: Real-World Tests on 10 Devices with OptiWize SDWAN — OptiWize