Dokku vs. Google Compute Engine

If you have your own in-house servers or have a cloud server with the freedom to configure any PaaS that you want, then Dokku is for you. So far, I have never run into a scenario where Dokku was not able to fit my needs, after deploying many different types of applications with varying frameworks, languages, and connected services. If you already have a PaaS available, however, Dokku may not be for you.

Incentivized

Google Compute Engine is so easy to implement and run. It doesn't require much knowledge to build an app since they provide multiple options to choose from with their prebuilt sample list. We can easily make customization on any website app we built for our client according to their needs and make changes if required.

Incentivized

• Dockerfile deployments are an incredibly simple and straightforward way to spin up applications.
• The docker-options plugin allows endless direct configuration of options passed to different docker lifecycle stages.

Incentivized

• Scaling - whether it's traffic spikes or just steady growth, Google Compute Engine's auto-scaling makes sure we've got the compute power we need without any manual juggling acts
• Load balancing - Keeping things smooth with that load balancing across multiple VMs, so our users don't have to deal with slow load times or downtime even when things get crazy busy
• Customizability - Mix and match configs for CPU, RAM, storage and whatnot to suit our specific app needs

Incentivized

• Dokku might overreach on its assumptions about how you want to host your app, using its own VHOSTS plugin by default. This may be a pro or a con depending on what you want.
• Proxy port configuration can be somewhat difficult, with Dokku resetting any pre-deployment configuration to "smart" defaults on the first deployment.

Incentivized

• Built-in monitoring via Stackdriver is quite expensive for what it provides.
• Initially provided quotas (ie. max compute units one can use) are very low and it took several requests to get an appropriate amount.
• Support on GCE is limited to their knowledge base and forums. For more hands-on support provided by Google, you must pay for their Premium services.

Incentivized

Overall services are good to go. Received good feedback from users. Have regional server locations. It has free extra service included.

Incentivized

Having interacted with several cloud services, GCE stands out to me as more usable than most. The naming and locating of features is a little more intuitive than most I've interacted with, and hinting is also quite helpful. Getting staff up to speed has proven to be overall less painful than others.

Incentivized

There were some very few cases where we had an outage but mostly it did not last more than 10 minutes. And if there were any maintenance, we were notified beforehand. So did not have any major problem with Google Compute Engine's availability.

GCE is a performant infrastructure. The GCE usage reports are readily available and the reports load time is insignificant. GCE exposes various REST API which allows the third part system to integrate with GCE. Many observability software like newRelic integrates with the GCE to offer monitoring of GCE

Incentivized

Dokku is incredibly well documented and also takes advantage of the documentation and community of Heroku, being essentially the same thing, but in the case that you encounter an actual bug or issue, it can take time for a fix to make it into a new version. Since you manage Dokku yourself, there is no support team to call for help.

Incentivized

• The documentation needs to be better for intermediate users - There are first steps that one can easily follow, but after that, the documentation is often spotty or not in a form where one can follow the steps and accomplish the task. Also, the documentation and the product often go out of sync, where the commands from the documentation do not work with the current version of the product.
• Google support was great and their presence on site was very helpful in dealing with various issues.

Incentivized

Dokku is essentially the same thing as Heroku, except maintained by you on the platform of your choice. If you have any experience using Heroku, then you will feel right at home using Dokku. Dokku is an entirely free to use PaaS, requiring only that you have a platform to deploy it on, making it far more versatile than Heroku in my opinion.

Incentivized

Google Compute Engine provides a one stop solution for all the complex features and the UI is better than Amazon's EC2 and Azure Machine Learning for ease of usability. It's always good to have an eco-system of products from Google as it's one of the most used search engine and IoT services provider, which helps with ease of integration and updates in the future.

Incentivized

• Since there's no investment besides a little time to set up, the return has been overwhelmingly positive.
• Deploying a new app is incredibly quick and easy.

Incentivized

• With Google Compute we don't have the overhead of managing our own data centers reducing costs and reducing the staff needed to manage systems.
• As I said earlier, Google's costs are ~1/2 of AWS, so we are able to see a ROI much faster.

Incentivized