Planning worksheets

Identify the primary use cases that you are trying to solve with this product. These use cases must determine the type of data you need to collect and the users that will interact with the system.

High-level deployment planning must help you answer the following questions:

1. What kind of data do I need to collect and how much data will be involved? (See Planning sizing and scalability.)
2. How critical is the data?
   If the data that you want to collect is critical, you might want to consider implementing data collection redundancy.
3. What type of collection mechanism will I be using in my environment? (use Collection Agent or not) (See Agent types.)
4. What kind of tasks will users be performing on a daily basis with the help of this product, and how many users will be interacting with the product? (See Planning for users.)
5. How will I manage or administer the product? (See Planning for product administration.)
6. Are there any security considerations that I must keep in mind? (See Security.)
7. What kind of deployment is suitable for my needs? (See Deployment use cases and best practices and Indicators for scaling.)

The answers to the preceding questions can help you get an idea about the architecture and the number of hosts required to support it.

Review system requirements for:

• TrueSight IT Data Analytics server (See System requirements.)
• Collection Agent (if PATROL for IT Data Analytics is to be used; see  System requirements for PATROL for IT Data Analytics 2.7 .)

Consider your sizing and scaling needs for the next 6 to 12 months before finalizing specific hosts or virtual machines (VMs).

Identify the external dependencies for using the product. For example, answer the following questions:

• Will you be integrating with a supported product? If yes, coordinate and have a plan to work with the respective team on the integration points. For more information, see Integrating.

• Are there any custom collectors required to satisfy the use cases? For example, do you need to create some script collectors to feed the required data into the product? For more information, see Collecting data into the system.
• Are there any custom collectors required to satisfy the use cases? For example, do you need to create some scripts to be used while creating the notification, so that notifications are sent via a script? For more information, see Setting up notifications to create alerts or reports.
• Do you plan to integrate with TrueSight Presentation Server? If yes, you need to first configure details of the Atrium Single Sign-On server on IT Data Analytics. For more information, see User authentication options.

These scenarios must be identified and factored in to the deployment-plan timeline.

• Determine the number of users that will be actively using the product on a daily basis.

• Categorize the users so that you can clearly identify the impact and sizing implications of those users.

  Note: Notifications act as virtual users from a sizing and impact perspective. Every notification triggered automatically is equivalent to a user performing a manual search, identifying a problem, and triggering a notification.

  For more information, see Sizing and scalability considerations.

Determine if all users must be given the the same privileges and access to the same data:

• If yes, your planning for user creation and management is simplified.
• If no, you must identify the different types of user roles that you need to configure before deployment.

For more information, see User roles and permissions.

User group information and user credentials are stored in the Configuration Database.

You can authenticate into the product by using the IT Data Analytics mechanism or by using Atrium Single Sign-On. For more information, see User authentication options.

Identify the different types of data that you want to collect and index into the product. This step is important, because all other steps in the planning process are dependent on it. This step might also help you decide if you want to use the PATROL for IT Data Analytics or a remote collection mechanism; for example, Windows Share or Secure Shell (SSH). 

Perform the following steps:

1. Identify all of the types of data that you require for troubleshooting and correlation use cases.
2. Determine where the data resides and how will you access it (also identify the type of data collector appropriate for your environment).
3. Prioritize the type of data you want to collect, and create a plan for incrementally fetching different types of data over a particular timeline. Identify all data that might be important to search before starting your troubleshooting and trying out the correlation use cases that you have in mind.
   

For more information, see Setting up data collection.

After identifying the types of data to be collected, you can identify the expected volume of data that is expected to be collected per day. This step is critical for proper sizing of the product components. Consider how your data needs might change in the next 6-12 months and factor that into your deployment plans.

For more information, see Sizing drivers and their impact.

Determining the number of concurrent users and types of searches or notifications that can be run on an hourly or daily basis is important for proper sizing. Likewise, understanding the types of users that might operate in the near future (6 to 12 months) is also important for proper deployment planning.

For more information, see Planning for users.

If you plan to change the default data retention period under Administration > System Settings, then you need to factor this into your sizing and scalability plans.

For more information, see the following links:

• Horizontal scaling recommendations
• Long-term data retention recommendations

Although it can be difficult to get a detailed understanding of what you want do with the data during the planning phase, you must consider the number of fields that will be extracted from the various types of data collected. Fields provide a layer of structure on the top of the semistructured data that you collect and can be an important part of search or correlation use cases. At this stage, be aware of this point and try to estimate the number of fields based on your expected use cases. 

For more information about accounting for fields, see Variables that impact product performance.

Notifications are critical if you are integrating with other BMC products (for example, ProactiveNet or TrueSight Presentation Server). You can use these notifications to send alerts to the integrated products. You must understand the notification requirements as early as possible to be able to consider them for your sizing and scalability planning. Similarly, you must determine the impact of the number of concurrent users.

For more information, see Variables that impact product performance.

• Determine if user creation needs to be done via IT Data Analytics or Atrium Single Sign-On. For more information, see User authentication options.
• Determine if users created need to be coordinated with any other product's user names or password-management system.
• Consider roles and permissions that you want to grant to users; simplifying the number of roles and variations can make management easier.

Data collection management

(with or without using PATROL for IT Data Analytics)

If you want to use PATROL for IT Data Analytics for data collection, your planning process must incorporate configuration needs related to the following systems:

• ProactiveNet
• PATROL for IT Data Analytics

You must understand the interaction between this product and the preceding systems before deciding on your deployment architecture. For more information, see Agent types.

For more information about effectively managing your data collection, see Setting up data collection.

Data Collection Management

(remote)

You can use various data collectors for remote collection of data, such as:

• Monitor file over Windows share
• Monitor file over SSH
• Receive over TCP/UDP

For more information, see Collecting data into the system.

If you want to use any of these data collectors, you must understand the implications on your deployment architecture. Also, in some cases, you might want to consider your department or company policies to determine your data-collection management. For example, if your company has strict firewall policies whereby the managed nodes cannot connect with the server, you might want to use one of the remote data collectors listed previously. Accounting for such factors at the beginning of the planning process can help smooth your actual deployment later.

For more information about effectively managing your data collection, see Setting up data collection.

The product might require periodic maintenance or administrative attention. The following duties and tasks are applicable for an administrator:

• Ensure that data collectors are synchronized with your IT environment.
• Ensure that users and user groups are synchronized and up to date.
• Ensure that notification management is synchronized with the integrated products (for example, ProactiveNet) or email recipients.
• Manage global settings such as data retention, timeout settings, and so on.
• Number of views can be a factor that impacts product performance.
• Decide a backup strategy to help protect against disasters that might result in data loss. For more information, see Backing up and restoring data.

For more information, see Planning.

All external connections to and from the IT Data Analytics Server can be enabled via Secure Sockets Layer (SSL). During deployment planning, you must decide if you want to use a secured connection to enable you to assess configurations and performance impact.

For more information, see Configuring a secured connection.