Data warehouses are meant to keep huge data, often collected over a number of years. This often demands allocation of the right amount of space at the beginning when you set up your data warehouse. But often wrong estimation or forecasting of the space might result in your data warehouse failing over a period as data gets accumulated in it, thus, failing your application. This article presents guidelines about sizing BMC Decision Support – Network Automation data warehouse correctly so that it does not run out of space over period.
This article would mainly help BMC Decision Support - Network Automation administrators during new deployments.
The most fundamental factor that determines the space requirements is how many years do you want to retain the data for in the data warehouse. Once you decide the time period, next step is get the number of network devices managed by BMC Network Automation. Once these two factors are determined, you can use the following guidelines to determine the space required by the data warehouse.
Assuming BMC Network Automation is managing 25000 devices, then here are space requirements for it:
Size of Work and Master databases does not grow with the increase in the number of devices managed. So, the space requirement does not change.
Note: Size of DWH changes as per ETL frequency because with each ETL, DWH stores snapshot of compliance and drift status for the entire network and other information.
For example, If you are managing 100,00 devices through BMC Network Automation and you want to store the data for 5 years with ETL running every 24 hours, then the space requirements would be as follows:
DWH Size – [ 10 GB * (100,00/250,000) ] * 5 = 20 GB Work DB Size – 3 GB Master DB Size – 2 GB
So, the space required would be 25 GB (20 + 3 + 2). Hope you find this article to be useful while you are planning your new deployment for BMC Decision Support – Network Automation and would avoid your data warehouse from running out of space as time passes. Its always better to be proactive then reactive.