I’ve always heard people say FTPS is faster than SFTP.
Turns out, that’s not really the whole story.
When I first read this new benchmark, it flipped a few assumptions on their head: SFTP vs. FTPS benchmarks: file transfer speed comparison 2025.
What made it interesting wasn’t just the numbers, but how different things affected the outcome. For example:
- With lots of small files, SFTP actually handled the workload better.
- When it came to one big file, FTPS had the upper hand and sometimes by a wide margin.
- Switching the client from curl to LFTP totally changed the picture. SFTP suddenly looked way better, while FTPS lost some of its shine.
- Add in cross-region transfers with more latency, and the balance shifted again.
- So the whole “FTPS is faster” line doesn’t really hold up. It depends on the workload, the client (to a great extent), and even the distance.
What I’d really like to see next are tests with parallel transfers, since most real-world setups don’t move files one at a time. Also curious how things would look with other clients or libraries, or at larger scales like 10GB batches.
But that’s just me. I’d love to hear from others:
- What use cases do you run into?
- Any nuances with certain scenarios or clients?
- Do you mostly deal with tons of small files, or big chunky ones?
- Have you run your own SFTP vs FTPS tests, and did you see similar results?
- Anything specific you’d want to see tested in future benchmarks?
- Other protocols you need to see up against SFTP/FTPS? Thoughts?
LA.
Top comments (0)