Alright, so the entire internet seems to think the zero-trust/ZTNA conversation starts and ends with Twingate or Tailscale. I get it, they're good. But come on, the market's bigger than that.
I've been poking at Banyan for a potential enterprise POC. The policy-as-code angle is genuinely interesting for our data pipeline teams, but the pricing model is making our finance people do that slow blink. You know the one.
So I'm looking for alternatives that actually bring something *different* to the table. Not just another WireGuard wrapper with a pretty dashboard. I'm talking about real architectural differences: maybe deeper SaaS app integration, a unique take on device trust, or a billing model that doesn't assume I have infinite budget for "connectors."
What's out there that's actually worth a look? I'm specifically *not* interested in "just use Twingate" or "Tailscale is the answer." Assume I've already read those threads. I want the weird, the niche, the over-engineered, the surprisingly elegant. The stuff that makes you go "huh, that's an interesting choice."
Hit me with the obscure contenders or the established players who are doing ZTNA in a way that doesn't feel like a clone. Bonus points if they have a strong story for ephemeral access or tie into a broader data/analytics stack.
just sayin'
Data over dogma.
Yeah, the finance slow blink is real. We looked at Banyan too and had the same reaction.
For something with a different architectural take, maybe check out NetFoundry? Their whole thing is built on application-specific networking, not just tunneling. It felt less about "connectors" and more about tying access directly to the app identity, which could align with your data pipeline use case.
Have you run across them yet? Their pricing seemed less opaque, at least.
NetFoundry's app-specific approach is a solid alternative for that use case, you're right.
Just watch the data transfer costs if your pipelines move serious volume. Their model is clearer on user/seat pricing, but the egress can sneak up on you. It's the classic "we saved on licenses but got hosed on bandwidth" swap.
Any other contenders you evaluated that had a similar 'no-tunnel' architecture? I'm curious if anyone else is attacking it from that angle.
- elle
I was in a similar spot last year. If the policy-as-code angle is key, have you looked at OpenZiti? It's the open source project behind NetFoundry's commercial offering, and you can self-host it. You get that same app-embedded, no-tunnel architecture, but with more control over the cost variables (like egress). The trade-off is you're managing the control plane yourself.
For a truly different billing model, check out Perimeter 81. They're more SASE-focused, but their pricing is per user, flat-rate, which our finance folks loved. It bundles FWaaS and SWG, so it might be overkill if you only need ZTNA, but the simplicity was a win.
Both avoid the "connector tax" pretty effectively. OpenZiti's learning curve is steeper, but if your team is already pipeline/Infra-as-Code oriented, it might fit.
Cloud cost nerd. No, I don't use Reserved Instances.
Thanks for this. When you mention managing the control plane yourself for OpenZiti, is the learning curve more about the initial deployment or the day-to-day policy updates?
The flat-rate pricing for Perimeter 81 sounds really attractive for budgeting. Do you know if they lock you into a long contract for that rate, or is it month-to-month?
Yeah, that's a great breakdown of the trade-off. For the control plane question, I found the day-to-day policy updates easier once we got our heads around the fabric model. The initial deployment was the real hump, especially figuring out where to host the controllers for latency.
I've been reading about Perimeter 81 too. That flat per-user rate is super clean for planning. Does their flat bundle include any kind of logging or audit features that you'd normally pay extra for? That seems like the kind of thing that could swing the value.
rookie
That's a really good question about the logging. From what I saw in their sales deck, the standard flat-rate package includes basic event logging, but the full audit trail and compliance reporting features are in their higher-tier plan. It's a common split.
Your point about latency for the OpenZiti controllers is spot on. We ended up deploying ours on a regional cloud provider closer to our primary workloads, which made a noticeable difference in handshake times compared to our first attempt on a central cluster. It added a step, but it was worth it.
The latency fix is a nice hack. You still traded vendor overhead for operational overhead, though.
The logging split you mentioned is a classic. "Basic event logging" usually means "you can see something happened, but good luck figuring out who or what without the upgrade."
What's the breakpoint on user count where their higher-tier plan becomes cheaper than just paying the egress elsewhere? Bet their sales team gets quiet on that one.
-- old school
OpenZiti is the obvious one for "policy as-code" and a different architecture, but you have to want to run the plane.
The truly weird niche contender I've seen is Nebula from Slack. It's raw, it's just crypto and config files, and there's no vendor at all. You'll build your own trust model from scratch. It's the "over-engineered and surprisingly elegant" box ticker, but you trade all vendor overhead for all operational overhead.
And yeah, avoid anything that charges per "connector." That model is just a tax on having more than one service.
Beep boop. Show me the data.