This article will introduce you to the MapTiler Cluster. It is going to show you the basic steps necessary for running/using it and it will also show you several handy tools that will make the work with MapTiler Cluster easier and more convenient for you.
Contact our specialists in case you are not sure whether MapTilec Cluster is the right solution for you.
About MapTiler Cluster
A simple rendering of a dataset on the Google Cloud Platform
MapTiler Cluster v1.0 is currently available as a service on G Suite Marketplace. Version v2.0 will be available for on-premise deployment on any other public or private cloud.
Launch MapTiler Cluster on G Suite Marketplace
For launching the MapTiler Cluster, go to G Suite Marketplace and search for MapTiler Cluster. Open the application and you will see the welcome page with information about it. Click on the Launch on computer engine button.
In the next window, you set parameters of the machine(s) which will process your data. You need to name the deployment, the instance, and set the parameters of your machine. The default value is 4 CPUs and 15GB memory: if you need to increase it, click on the Customize link and drag sliders with the number of CPUs and memory size. You can also extend memory or choose a machine type.
Selecting a zone means choosing where your data will be processed and stored. There is also a difference in prizes between zones.
The number of local SSD determines how many hard drives will be used for data storage.
Before you click on the Deploy button, you also need to tick the I accept the GCP Marketplace Terms of Service checkbox. At this moment, you start paying (small amount) for the deployment of your machine.
After you click on Deploy, a new machine will be launched. It takes some time; in the meanwhile, you will be informed about the deployment of particular components. When ready, you will be informed by a notice "maptiler-cluster has been deployed".
On the right side, you can see information about your machine: Admin username, Admin password, and VM IP address of the machine.
Simply click on Visit the site button, put there your Admin username and password to start with the wizard. You may be warned about a self-signed SSL certificates and accept this unsafety to proceed to the site. You will be able to change the SSL certificate for HTTPS later.
If you want to stop the machine at this moment, click on the Delete button.
MapTiler Cluster wizard
On the given address, after writing the given admin username and password, you should see the MapTiler Cluster graphical wizard. On the welcome page, just click on the Run Setup Wizard button.
First, you need to specify the input for your map. This should be a bucket on Google Cloud Storage. For testing purposes, you can use mtc-sample-data bucket, which offers aerial imagery of the San Diego area. Read this official document for learning about Google Cloud Storage buckets and how to create them. Or you can read our steps on how to transfer your data into Google Cloud Storage.
The Advanced options link allows Filename filter, adding MapTiler arguments and File arguments script. All these functionalities are described in MapTiler Cluster manual. All parameters can be undone by clicking on the Reset button.
When ready, click on the Scan button.
During scanning your files, you can visually see the whole area you are going to render with information about each and every file which will be on your final map. If everything seems to be all right, click on the Next button.
In the following step, you set the output for your map. Currently, Google Cloud Storage is supported. Create a new bucket by writing a unique name.
Zoom is calculated by MapTiler automatically, however, you can change these settings by writing preferred min and max zoom level.
As an Output tile format, there are several options to choose from. Read more in the manual page for MapTiler Engine.
In Advanced options, you can set MapTiler arguments for rendering, an URL of Custom watermark, Project cutline, and File preprocess script. For more information, read the MapTiler Cluster manual.
When ready, click on the Next button.
Now you can set up your Cluster. First, insert your MapTiler Cluster license key. Without a license key, the map will be rendered with our watermark (and you will be asked to confirm the warning in a pop-up window). You can purchase the license key here.
Set the Number of virtual machines (workers) and the Region where they are running. You can specify a different number of workers in various regions depending on actual availability or your preference.
Detailed information link will show you info about required performance for various rendering options.
Clicking on the Render button will start the actual map rendering process.
The system will initialize the cluster (you can see the advance on the progress bar) and then the rendering starts showing visually which part of your maps is being rendered, what is the overall progress and the progress on each part. There are two graphs showing the speed of MapTiler in rendered tiles per second and the other one showing how fast are the tiles being uploaded to the bucket (again, in tiles per second).
When the rendering is finished, it will take some time until the whole process is finalized (stopping workers, preparing overview, ...). When finished, you will be automatically redirected to the final map.
Now you can finally view your map. All images are tiled and ready to be served from Google Cloud Storage.
Admin wizard page
If you would like to change the admin password or upload your own SSL/TLS certificates go to https://your-master-instane-IP/admin page.
After you are done with your rendering, it is advised to delete your deployment, because you are being charged hourly for running Google instances on the Google Compute Engine platform. In the picture below you can see where to delete your deployment. Go to the deployment page at the top click on the Delete button, it would take a few seconds after the deployment is deleted. Deleting deployment does not have any effects on your Storage buckets so you would not lose your data after the deployment is deleted.