New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
OVH vs. Hetzner Cloud Network Quality
Hey all! I would like to get your input on OVH vs. Hetzner Cloud network quality in general for the following locations:
OVH Limburg compared to Hetzner Nürnberg
OVH Vint Hill compared to Hetzner Ashburn
OVH Hillsboro compared to Hetzner Hillsboro
OVH Singapore compared to Hetzner Singapore
Any input would be awesome, I can see that Hetzner has only few transit providers and peerings in their non German locations, but in general I would like to get some input from real users with experience in those countries.
Thanks ![]()
Thanked by 1riomartin


Comments
OVH Singapore, as I located in Indonesia, the things to monitor was their peer connections to SGIX or Singapore Global Switch. Because many ISPs from here got peers connection in those two locations. But badly, OVH Singapore seems don't care. They forced route connections to their main link to USA. What is benefit of placing servers in Singapore if the route must be forced to USA before reaching Singapore. I am also waiting for Hetzner Singapore reviews to make sure every routing goes smoothly before decide to step in to their service.
Not comparing. Their network seems flawless so far. I've deployed my HA cluster application across three Providers.
They do a tcp heartbeat every 30ms each other (Raft Algorithm implementation) and never broken.
1) OVH Limburg (2 nodes) <-> Hetzner Nurnberg (1 nodes) <-> LET Providers (2 nodes)
2) LET Providers(2 nodes) <-> Hetzner Falkenstein (2 nodes) <-> One node will be added.
3) OVH VintHill (1 node) <-> Hetzner Ashburn (2 nodes) <-> LET Providers (2 nodes)
I'm curious, how do you determine all these connections?
Here is the NWS of a small idler (Hetzner, Nuremberg) that currently only runs a FastAPI + Next.js beta site with very limited traffic.
I have been monitoring it with HetrixTools since October 7, 2024, and there hasn't been any downtime except for restarts initiated by me.
I often found that OVH likes to route via their own network vs IXs
OVH Limburg compared to Hetzner Nürnberg
They are both solid in my opinion.
OVH Vint Hill compared to Hetzner Ashburn
I haven’t tried Hetzner enough but I like Vint Hill over Hillsboro. Connectivity seems to be better.
OVH Hillsboro compared to Hetzner Hillsboro
New location for both so.. improvements are needed.
OVH Singapore compared to Hetzner Singapore
Haven’t really tested Hetzner but OVH Singapore isn’t that good
Check Smokeping on these locations you'll find some interesting peering results. For example from Hilsboro to Vancouver on OVH some traffic is routed through Illinois, resulting in 70ms ping for what really should be a <10ms ping.
I think this is the same issue I am facing with their node in Singapore
)
From Jakarta to Singapore usually 10-12ms
but goes over 100ms, to Japan and US
and even worse, both traceroute showing different path, not the same
So in General OVH seems worse? Even though their network looks more sophisticated than simply getting transit from 2-3 providers as hetzner does. Interesting.