TigerGraph Cloud offers different choices when it comes to instance types and disk sizes. This document helps you find the right instance type and disk size for your workload.
Larger graphs require more RAM. Higher performance calls for both more CPUs and more RAM.
Our pricing table gives initial recommendations for which instance type to start with if you know how much data you will be loading into the graph. We call the input data your raw data. TigerGraph reorganizes your data into a graph, encoding and compressing it. We assume that your stored graph will be about 70% of the size of your raw data. This is a conservative estimate; it's often smaller.
You also need RAM for your graph querying and computation. Workloads vary considerably. In the pricing table, we recommend that your total RAM be about 1.5 times the maximum amount of raw data you think you will load.
Disk storage refers to the root disk attached to your instance that holds the following files:
OS installation files
TigerGraph database installation files
TigerGraph Graph storage
Other data files uploaded to the instance
Any output files generated by querying the database
Logs generated while running TigerGraph
When you choosing the size of the disk, consider the size of the data for all the above installations and files.
To ensure adequate disk storage, TigerGraph Cloud mandates that any disk size of a solution must be at least 3 times the size of its instance's RAM. For example, if your solution instance has 64 GB of RAM, then the disk size must be higher than 192 GB.
TigerGraph Cloud offers three cloud platform options: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP). The choice of cloud providers and regions affects the pricing and network latency of your solutions. The region indicates the physical location of your solution.
Each cloud provider has different offerings for instance type, disk storage, and backup storage. At the moment, TigerGraph Cloud only supports free-tier clusters on AWS. You may choose to deploy free-tier clusters on any region in AWS. For more information on the options of each cloud provider, see:
When you stop a solution, you shut down the virtual machine instance. The TigerGraph database is frozen in its current state. Billing for that machine instance also stops. Scheduled backups also stop, but the backup copies are kept in EFS on AWS, Azure Files on Azure, and GCP Filestore on GCP. You will still incur charges for disk storage and backups when a solution is stopped. You can start the solution again. After a warm-up period, the solution will return to its previous state.
Below are the steps to stop a solution
Find the solution you want to stop and click on the Solution Operations icon and click Stop.
You can restart a stopped solution at any time. After a warm-up period, the solution will return to the state it was in when it was stopped.
Below are the steps to restart a solution:
By terminating a solution, you deprovision the virtual machine and the attached disk storage. The solution is deleted according to the cloud platform's policies and cannot be recovered, so is everything in the attached disk storage.
Below are the steps to terminating a solution.
Find the solution you want to stop and click on the Solution Operations icon and click Terminate.
After a solution has been terminated, it can be archived so that it no longer shows up on the default view in the list of solutions.
To archive a solution, click on the solution row in My Solutions to expand it and click the Archived toggle.
To view archived solutions, click the settings icon to the right of the Actions column and toggle on Show Archived in the Display Settings.
Find the stopped solution, and click the Solution Operations icon , and in the dropdown list, click Start.
In this tutorial, you will learn how to provision a new TigerGraph Cloud solution.
After logging into your TigerGraph Cloud account, click My Solutions on the left side menu to go to the My Solutions page.
On the My Solutions page, click Create Solution.
Not all TigerGraph versions that are offered on-prem are offered on TigerGraph Cloud. In the dropdown list, choose the version of TigerGraph you want to run.
TigerGraph Cloud Starter Kits are built with sample graph data schema, dataset, and queries focused on a specific use case to help you get started quickly. Most of our starter kits have a demo video that walks you through the schema and the queries built for its use case. If you wish to proceed with your own schema and data, select Blank.
After choosing the starter kit, scroll down to the bottom and click Next.
The choice of cloud provider affects the configuration options for the available instance types and disk sizes, network latency for accessing your solution, the geographic location of your solution, and the cost of running your solutions. For more information, see Cloud Providers and Regions.
Each instance type indicates its vCPU and RAM specifications as well as the hourly rate for running the instance. For suggestions on choosing the right instance type and disk storage, see Sizing and Scaling.
Region reflects the physical location of the server running your solutions. It also affects the network latency for accessing your solution.
Disk storage refers to the root disk that is attached to your instance. If you need help choosing the right disk storage, see Sizing and Scaling.
To ensure your solution has enough disk storage, TigerGraph mandates that any solution's disk size must be at least 3 times the size of the solution instance's RAM. For example, if you choose an instance type that has a RAM of 64 GB, then the solution's disk size must be more than 192 GB.
In a High Availability (HA) cluster, the partitioning factor refers to the number of parts or components your graph data is split into, which also equals the number of instances that collectively store one copy of the full graph.
If you only need a single instance to store your entire graph, enter 1 as your partition factor.
The replication factor is the number of copies of data, each on a separate machine. TigerGraph Cloud currently supports a replication factor of up to three.
If you don't need additional copies of your data, enter 1 as your replication factor.
If you provision a cluster with a replication factor higher than 1:
If your solution is on AWS or GCP, the replicas will automatically be provisioned on different availability zones.
A load balancer would be automatically attached to your cluster to balance traffic between the replicas.
If your cluster is on GCP and it has a replication factor higher than 1, you won't be able to use ports 9000 and 14240 to access API endpoints. Instead, you can access the endpoints using HTTPS requests on port 443:
To access the endpoints that normally listen on port 9000, append/restpp
after the domain and before the endpoint. Since an HTTPS request already implies port 443, specifying the port is optional. However, you need to ensure that you are making HTTPS requests instead of HTTP requests.
Example: To access the /echo endpoint: curl "https://examplesolution.i.tgcloud.io/restpp/echo"
To access the endpoints that normally listen on port 14240, use the original endpoint address, but on port 443. Similar to port 9000 requests, specifying the port is optional.
Example: To access the health check endpoint on 14240:curl "https://examplesolution.i.tgcloud.io/api/ping"
Access to GraphStudio is not affected: opening the domain of the solution in a browser will directly take you to GraphStudio
Give your solution a name. A solution name can only contain alphanumeric characters, space, dash, and underscore and may be no longer than 20 characters.
Give your solution a tag, which helps you sort and identify your solutions. A solution tag may be no longer than 40 characters and can contain any character.
When a solution is provisioned, a default TigerGraph user tigergraph
is created in the installation process. This is the initial password for the default user.
This initial password allows you to log into the database when your solution is in the ready state. If you forget this password, you will need to terminate and recreate the solution.
Enter a unique subdomain for your solution. Only letters, numbers, and inner hyphen(-) are allowed in the subdomain. You can send HTTP requests to the REST endpoints on the domain once the solution is up and running. If you do not enter a subdomain, an automatically generated subdomain will be assigned to the solution.
Enter a description of what the solution is used for or anything else you would like to note about this solution.
After making sure all your settings are correct, click Submit to start your solution. Your solution should be ready after a brief warm-up period.
Solutions are fully-managed TigerGraph deployments. Here are some resources you can use to create, configure and manage your solutions:
A step-by-step guide to creating a new solution
Information on the different cloud providers and regions available on TigerGraph Cloud
Learn how to choose the appropriate instance type and disk size
Stop, Resume and Terminate a Solution
A step-by-step guide on how to stop, restart, and terminate a solution