Maintenance Notification:
On next Monday August 5th 2019 starting at 7am CEST we will conduct some maintenance. You might encounter some issues using the Exasol User Portal and Issue Tracker! We will restore the Exasol User Portal and Issue Tracker before 9am CEST on Monday August 5th 2019.

You are currently viewing an old documentation portal. This content will be removed on June 18th, 2019. Please view our current documentation portal at
Skip to end of metadata
Go to start of metadata

Once you have understood the concepts and architecture of Exasol (refer to the Overview for this information), you can start planning how the Exasol system will be implemented in your environment. During this phase, you make decisions that will have an impact on performance and scalability of the Exasol database, and the effect that hardware or network failures will have on the system and your operational activities. 

Hardware Sizing

Hardware sizing refers to the processing power, storage space and the hardware configuration of the environment in which Exasol will run. A sizing estimation takes several input factors into account, such as the following:

  • Expected data volume to be stored in the cluster
    This is the uncompressed data volume of the data you will store in the cluster. It is perhaps the most important factor, because it impacts both storage and performance requirements. 

  • Performance requirements
    The Exasol database does not require all the data to be put into RAM. If queries require reading data from disks, the execution times start varying. To achieve good performance, you should provide the database with enough RAM (typically 10% of uncompressed data) to hold your active data. However, you may need to increase this value in some cases. For example, if there is a web-application that accesses a certain amount of data, and execution times are expected to be constant, you might need to consider a higher value for RAM.

  • Redundancy requirements
    Exasol supports storing data redundantly to prevent data loss in the event of a hardware failure, such as the failure of a node in a cluster. If you opt for redundancy, you effectively increase your storage requirements. For example, if you choose a redundancy of '2' (meaning that each segment of data is stored twice) your storage requirements are doubled.

  • Planned backup strategy
    If data backups are stored in the cluster (as opposed to a remote server, for example), storage requirements are increased. Other factors like backup redundancy and backup frequency also affect the sizing estimation. 

You probably already have some knowledge of your current requirements. However, your sizing estimation should go one step beyond only considering current factors; you also need to consider future growth of data volumes or operational requirements. The hardware configuration that you opt for should encompass your current needs as well as allow for potential growth in the future. 


Exasol provides a wizard to help you through the sizing estimation. For more information on sizing and on using the wizard, refer to: Hardware Sizing

Planning Topics

Network Planning

Network planning should be done once you have completed a hardware sizing estimation. The reason for this is that during the sizing step, you will define your expectations of storage size, performance needs, as well as your backup strategy. These aspects will determine the network strategy, such as number of network interfaces you need, connection speeds and other parameters. For more information on planning the network environment for Exasol, refer to Network Planning.

Business Continuity Planning

A business continuity plan provides a strategy for recovering from an outage or failure. The nature of the plan has an impact on the hardware and network requirements for your environment, and is therefore important to consider during the planning phase. For more information, refer to Business Continuity Planning.

  • No labels