Next: Creating Growth Patterns (2 of 2)



The Basics

Welcome to PlanForCloud. This getting started guide will go through how to create cloud cost forecasts and will give details on how to create and use patterns to model growth. By the end of this guide, you will be able to re-create the example 3-tier web app deployment which is created for you when you first log in.

When you log in as a guest or sign up and log in, you will see the dashboard as well as the example deployment:

The Dashboard
  • Deployments: a deployment represents a groups of servers, storage and databases. Each deployment has its own cost report. You can create deployments to represent different scenarios (e.g. using different types of servers or patterns) and get detailed cost reports for each.

  • Patterns: one of the key benefits of using the cloud is elasticity (the idea that your system can scale-up or scale-down to how much resources it needs dynamically). This means that you can create deployments that do not have fixed monthly costs. To generate more accurate cost forecasts, you can create patterns and attach them to various numbers in your deployments. For example, an online shopping system can use a pattern to model how the number of web servers it needs changes during the busy shopping season. Patterns are optional, so you can start without them.

  • Cloud cost forecast: each deployment has its own cloud cost forecast report that shows the estimated cost of that deployment over three years from the date of report generation. The report also shows a breakdown of the costs and a table of the monthly costs.

Please note the following considerations when using PlanForCloud:
  • Cloud providers charge in different currencies. In order for your cost reports to be shown in your chosen currency, we use Google Money to get live exchange rates at the time of report creation.
  • The 'General Availability' (GA) prices of Windows Azure are used in the simulations. Microsoft are using discounted 'preview' prices for now but will probably soon switch to GA, so it's best to assume GA prices in your cost forecasts to avoid any surprises.
  • AWS Spot instances are currently not supported.

Creating Cloud Cost Forecasts


Creating a deployment

It couldn't be simpler to create a deployment - enter a name for your deployment and click the Create Deployment button. The new deployment will automatically open for you. We will name this example deployment '3-tier web app': Create Deployment


You can also import a snapshot of your AWS account using the Import button. We only require read only credentials: Create Deployment from AWS

Creating a server

Click the 'Add Server' button
The following page will appear which enables you to filter the server selection by cloud provider, CPU count and RAM (GB).
Create Servers
Click on one of the server rows to select that server and enter its usage and quantity details.
Once done, click the 'Add Server' button. This will add the selected server to your deployment.

You can add more servers or edit existing ones from the 'Servers' tab within a deployment.


Creating a storage unit

Click on the 'Storage' tab within a deployment. Then click the 'Add Storage' button.
The following page will appear which enables you to filter the storage selection by cloud provider, region and text search (e.g. 'Glacier').
Create Storage
Click on one of the storage rows to select that storage. You can name your storage, define the quantity of these storage units required, the size of each unit (in GB), the read and write requests to each storage unit per month (if unknown, leave as 0).
Once done, click the 'Add Storage' button. This will add the selected storage to your deployment.

You can add more storage units or edit existing ones from the 'Storage' tab within a deployment.


Creating a database

Click on the 'Databases' tab within a deployment. Then click the 'Add Database' button.
The following page will appear which enables you to filter the database selection by cloud provider, CPU count and RAM (GB).
Create Databases Click on one of the database rows to select that database. You can name your database, define the quantity of these databases, define their usage (e.g. 24 hours/day), their size and also transactions to the database per month (if unknown, leave as 0).
Once done, click the 'Add Database' button. This will add the selected database to your deployment.

You can add more databases or edit existing ones from the 'Databases' tab within a deployment.


Creating a data transfer link

You can model data transfers within a deployment, or from inside your deployment to external sources (e.g. users). Click on the 'Data Transfer' tab and select the source and the destination of the data transfer. For example, in the example 3-tier web app, to represent the data transferred between the users and the Load balancer, select:
Source: Remote Node: User
Destination: Server: Load balancer - HAProxy
Source to destination (GB): 80
Destination to source (GB): 480
Click Create Create Data Transfer


Create cloud cost forecast report

At any point in time, you can click on the See 3-Year Cost Report button of a deployment to see its cost forecast report. The example deployment (with patterns applied - which we will look at on the next page) creates a report which looks similar to this:
Cloud cost forecast from PlanForCloud

Next: Creating Growth Patterns (2 of 2)