Sizing and scalability considerations
TrueSight Capacity Optimization components are designed to enable you to scale the TrueSight Capacity Optimization infrastructure to manage environments with varying numbers of managed systems, data feeds, reports, users, and amounts of historical data.
Important
This topic provides general sizing and scalability guidelines. Contact the BMC Customer Support for your sizing requirements.
Sizing drivers
The following figure shows the major sizing drivers for each TrueSight Capacity Optimization component:
TrueSight Capacity Optimization sizing drivers
A major sizing driver for the TrueSight Capacity Optimization infrastructure is the required throughput for processing collected samples. One TrueSight Capacity Optimization unit refers to one million samples per day. For example, 20 million samples per day are also referred to as 20 TrueSight Capacity Optimization units.
An entity is a managed system or device whose performance data is collected, such as a physical stand-alone server, a partition on a server, a virtual server, a database instance, a storage element, or a network element.
The following table shows how to translate the number of managed entities, metrics, and time granularity of the samples into the number of samples per day.
Number of managed entities | Average number of metrics per entity | Time granularity of the samples | Samples for 24 hours |
|---|---|---|---|
| 1,250 | 40 | 1 hour | 1.25 million |
| 2,500 | 40 | 1 hour | 2.5 million |
| 5,000 | 40 | 1 hour | 5 million |
1,250 | 40 | 15 minutes | 5 million |
2,500 | 40 | 15 minutes | 10 million |
5,000 | 40 | 15 minutes | 20 million |
Note
The preceding table does not take into account business driver metrics. Estimate the number of business driver metric samples that have to be processed on a daily basis as 10% of the overall managed entity metric samples per day.
The following topics describe guidelines and recommendations for sizing each TrueSight Capacity Optimization component, based on average utilization scenarios. All sizing values are provided as general guidance.
Deployment size
The following table shows TrueSight Capacity Optimization deployment sizes according to these two major sizing drivers:
- number of managed entities
- number of samples to process
Small | Medium | Large | X-Large | XX-Large | |
|---|---|---|---|---|---|
<= 20 M rows | <= 40 M rows | <= 80 M rows | <= 160 M rows | <= 600 M rows | |
Oracle | 4 cores 8 GB RAM | 4 cores 16 GB RAM | 8 cores 32 GB RAM | 16 cores 64 GB RAM | 64 cores 128 RAM |
Postgres | 6 cores 12 GB RAM | 12 cores 24 GB RAM | 24 cores 48 GB RAM | N/A | N/A |
Data Hub1 | 4 cores 16 GB RAM | 4 cores 16 GB RAM | 4 cores 16 GB RAM | 8 cores 32 GB RAM | 32 cores 128 GB RAM |
1 - A deployment can have only one Data Hub.
- The table provides basic guidelines and does not represent the complete range of scalability options. Contact the BMC Customer Support for your sizing requirements.
- These deployments are based on average utilization and default aging policies.
- The current compatibility offering for the PostgreSQL database supports large environments that can handle <= 80 million records per day with the right resource allocation.
In a test environment, you can install the Remedy SSO Server and the Presentation Server on the same computer, but install the Application Server, Gateway Server, and Capacity Agent on separate computers. In a production environment, install each of the TrueSight Capacity Optimization components on a separate computer irrespective of the deployment size.
Comments
Log in or register to comment.