I am new to the field of API gateways, but am asked to draw a comparison between Tyk and Kong, in terms of features, ease of use, community, roadmap and pricing.
I have been able to find some info on Tyk’s website (and Kong`s ), but would like to have the views of more experienced people on that topic, especially the technical bit.
I hope that this question is a good fit in this forum.
PS: Out of honesty, I have to say I’ve posted the same question on Kong’s forum, hoping to get both sides of the story :-).
Support - According to Forrester, Tyk have the best support of all the API management platforms. Our average response time is less than an hour, and our roadmap is always build based on customer needs.
Kong Built on top of open resty which is built on top of Nginx. if there is a bug with openresty / nginx, Kong cannot help. Tyk however - entire stack owned and developed by Tyk. Meaning faster resolution of bugs.
To be objective, the performance of both products is likely to be very similar. When it comes to API Gateways, usually their performance is not an issue. For example, the single Tyk instance with auth and plugins turned on can give 2k requests per second, which is usually enough for any use case.
Fixed license cost per year. No limits or price considerations for the traffic, no. users, no. APIs, no. admins, no. gateways, no. DCs
Kong dashboard and portal available only for Enterprise customers, while Tyk one available even with free license. And in addition, based on all customer feedback, Tyk dashboard is the best from what you can find across other API gateways.
We are growing fast, we are sustainable, and playing a long game
Thank you for your answer.
I actually came across that web page yesterday.
I was able to verify that part about Kong relying exclusively on Lua for plugin development, whereas Tyk was less restricted. Which is a good point.
I just thought that the author of the comment might not be 100% objective: there are surely good things to say about Kong
I’ve sent you a DM about the BBVA article above. It’s not a like-for-like comparison, as they not only used our “demo, strictly not for production” docker install, instead of using the “production docker” install. But they also tested the Full Tyk Enterprise stack (GUI, ANalytics visualisation, Analytics export, Developer Portal, API Designer etc) against just running Kong headless.