Anonymous user
Not applicable
2021-06-09
12:42 PM
0
1286
- Bookmark
- Subscribe
- Email to a Friend
- Printer Friendly Page
- Report Inappropriate Content
Link copied. Please paste this link to share this article on your social media post.
2021-06-09
12:42 PM
Overview of Data Archiving
Originally published on Geo SCADA Knowledge Base by Anonymous user | June 09, 2021 09:42 PM
📖 Home Back
The type and amount of data that is included in an archive is determined by:
Every archive is represented by a user-created archive item in ViewX. The configuration of each archive item defines which historic data and Event Journal messages are stored for that archive, and where the archived data is stored (see ClearSCADA Help). Typically, archive items are created for each server in the system. Each archive item is then configured to archive the data from all of the items in the database.
The amount of data that is included in an archive is dependent on the Keep Online For n Weeks time and Archive After n Weeks period, both of which are defined on the Server Configuration Tool. These time periods are explained in detail later in this article (see Archive Life Settings).
A server will never archive data that it has already archived previously.
Historic data and event messages are kept in the database for a specified number of weeks after which they are deleted. The total amount of time for which historic data and event messages are stored is referred to as the Keep Online For n Weeks time. To avoid losing data, you should ensure that archiving takes place before the end of the Keep Online For n Weeks time. Note that historic data modifications and annotations are not archived and are kept online indefinitely.
The Keep Online For time period is divided into two parts:
New data can be added to the data that falls within this period. The writeable period starts at the current time and extends backwards a specified number of weeks before the start of the current week. Data received during the writeable period cannot be archived (as it can still be updated with new values).
You can specify the number of weeks for the writeable period in the Archive after n Weeks settings in the Server Configuration Tool. (There are separate Archive After n Weeks settings for historic data and Event Journal data).
Data within this period is available for archiving. New data cannot be added to the data that falls within this period. Any data received from outstations etc., during the Read Only period is discarded.
The read only period starts at the end of the writeable period (where the end of the writeable period is the Archive After n Weeks number of weeks, see above) and ends at the end of the Keep Online For time period.
The Keep Online For n Weeks setting defines the total amount of time that the data remains in the database. When this time expires, the data is deleted. The total amount of time is defined as a number of weeks starting from the beginning of the current week.
The Archive After n Weeks setting defines the amount of time that must elapse before the data is available to be archived. This is specified as a number of weeks before the start of the current week. The Archive After n Weeks setting should be set to at least 1 week and must be less than the Keep Online For n Weeks period.
For example, if the data is kept online for 4 weeks and archived after 1 week, then data up to 1 week old can be added to the database, and a total of 4 weeks data is available at any given time. As both periods begin at the start of the current week, there is also additional data for the current week available online (the data from the start of the current week to the current time). So, by the end of the current week, there will be almost 5 weeks of data online. There will be exactly 3 weeks worth of data available to be archived at any given time, however the server will not archive data that it has already archived.
It is possible to set the Archive After n Weeks period to 0, but we do not recommend you do this as it is likely to cause loss of data if there is an archive at the start of the week.
Data is archived whenever a user selects the Archive pick action (available on archive database items) or the Archive method is executed by a Schedule, Logic program etc. Typically, a Schedule is created that executes the Archive method on a regular basis, usually once per week. This ensures that archives are performed automatically. However, periodically, the archive disk(s) will need to be replaced with new blank media (before each archive disk becomes full).
On systems that contain more than one server (for example, hot-standby systems), each server can be set up to archive data separately. This provides redundancy and additional copies of the archived data (one per server). The archiving always takes place on the server that is connected to the archive media, irrespective of the server status. So, if a standby server is to archive data, the actual process of archiving will take place on the standby server (it does not need to become 'Main'). Only the instruction to initiate the archiving process is sent across the network from the Main server to the Standby server(s). Similarly, the standby server(s) only send archive status data back to the Main server over the network.
As each server keeps track of the data it has archived, a server will never archive data that it has already archived. When an archive is performed, the server only archives the data that has not yet been archived by that server (archiving is incremental).
You can view information about the data that has been archived by a particular server by using the Archive Volume Manager (right-click on the Server icon in the Windows task bar then select the Archive Volumes option from the context sensitive menu). The Archive Volume Manager displays a list of all the data that has been archived, and the list is organized according to the volumes (disks) on which the data is stored. You can also use it to create replacement volumes.
ClearSCADA can archive data and events to any type of disk supported by Windows, including CDs, DVDs, and hard disks. When archiving to CD or DVD, Third-Party packet writing software is required. Unfortunately, Third Party packet writing applications can be unreliable and can even cause data loss. For this reason, we recommend that you archive data and events to hard disk, DVD-RAM or any other disk based archiving technology as these are more robust and reliable than archiving directly to CD or DVD.
ClearSCADA will archive data and event messages to your chosen archive media whenever a user selects the Archive pick action or the Archive method is called by a Schedule, Logic program etc.
When the archive media (hard disk, DVD-RAM etc.) becomes full, ClearSCADA raises an alarm and cannot complete the current archive task. In this situation, you need to replace the full disk with a blank disk and perform another archive so that the remaining data is archived to the new disk.
We recommend that you always replace each disk before it becomes full. This makes it easier to search for archived data at a later date---if all the data for a particular period is archived on the same disk, it is easier to locate the data you need.
You may wish to backup the archived data, for example, for secure offsite storage. You can do this by using Windows or a Third-Party application to copy the contents of an archive disk to another disk. In a multi-server system such as a Hot-Standby system, each server can independently perform archives, so there are at least two copies of the archived data (one per server). This is dependent on the server configuration, for more information, see ClearSCADA Help.
Note: On multi-server systems, the actual archiving process takes place on the server that is connected to the archive media, irrespective of the server status. This means that standby servers can archive data and remain in standby mode - all they require from the Main server is the instruction to initiate the archiving process. In return, the standby server will send archive status data to the Main server when the archive process is complete.
Most users choose to periodically archive data, for example, once a week. It is also common practice to replace each archive disk at regular intervals, for example, every 6 months.
We always recommend that you :
Go: Home Back
What Data is Included in an Archive?
The type and amount of data that is included in an archive is determined by:
- The configuration of the archive database item
Every archive is represented by a user-created archive item in ViewX. The configuration of each archive item defines which historic data and Event Journal messages are stored for that archive, and where the archived data is stored (see ClearSCADA Help). Typically, archive items are created for each server in the system. Each archive item is then configured to archive the data from all of the items in the database.
- The archive life settings (made via the Server Configuration Tool)
The amount of data that is included in an archive is dependent on the Keep Online For n Weeks time and Archive After n Weeks period, both of which are defined on the Server Configuration Tool. These time periods are explained in detail later in this article (see Archive Life Settings).
A server will never archive data that it has already archived previously.
Archive Life Settings
Historic data and event messages are kept in the database for a specified number of weeks after which they are deleted. The total amount of time for which historic data and event messages are stored is referred to as the Keep Online For n Weeks time. To avoid losing data, you should ensure that archiving takes place before the end of the Keep Online For n Weeks time. Note that historic data modifications and annotations are not archived and are kept online indefinitely.
The Keep Online For time period is divided into two parts:
- Writeable
New data can be added to the data that falls within this period. The writeable period starts at the current time and extends backwards a specified number of weeks before the start of the current week. Data received during the writeable period cannot be archived (as it can still be updated with new values).
You can specify the number of weeks for the writeable period in the Archive after n Weeks settings in the Server Configuration Tool. (There are separate Archive After n Weeks settings for historic data and Event Journal data).
- Read-Only
Data within this period is available for archiving. New data cannot be added to the data that falls within this period. Any data received from outstations etc., during the Read Only period is discarded.
The read only period starts at the end of the writeable period (where the end of the writeable period is the Archive After n Weeks number of weeks, see above) and ends at the end of the Keep Online For time period.
The Keep Online For n Weeks setting defines the total amount of time that the data remains in the database. When this time expires, the data is deleted. The total amount of time is defined as a number of weeks starting from the beginning of the current week.
The Archive After n Weeks setting defines the amount of time that must elapse before the data is available to be archived. This is specified as a number of weeks before the start of the current week. The Archive After n Weeks setting should be set to at least 1 week and must be less than the Keep Online For n Weeks period.
For example, if the data is kept online for 4 weeks and archived after 1 week, then data up to 1 week old can be added to the database, and a total of 4 weeks data is available at any given time. As both periods begin at the start of the current week, there is also additional data for the current week available online (the data from the start of the current week to the current time). So, by the end of the current week, there will be almost 5 weeks of data online. There will be exactly 3 weeks worth of data available to be archived at any given time, however the server will not archive data that it has already archived.
It is possible to set the Archive After n Weeks period to 0, but we do not recommend you do this as it is likely to cause loss of data if there is an archive at the start of the week.
When is Data Archived and How do I Start the Archiving Process?
Data is archived whenever a user selects the Archive pick action (available on archive database items) or the Archive method is executed by a Schedule, Logic program etc. Typically, a Schedule is created that executes the Archive method on a regular basis, usually once per week. This ensures that archives are performed automatically. However, periodically, the archive disk(s) will need to be replaced with new blank media (before each archive disk becomes full).
Where Does Archiving Take Place?
On systems that contain more than one server (for example, hot-standby systems), each server can be set up to archive data separately. This provides redundancy and additional copies of the archived data (one per server). The archiving always takes place on the server that is connected to the archive media, irrespective of the server status. So, if a standby server is to archive data, the actual process of archiving will take place on the standby server (it does not need to become 'Main'). Only the instruction to initiate the archiving process is sent across the network from the Main server to the Standby server(s). Similarly, the standby server(s) only send archive status data back to the Main server over the network.
As each server keeps track of the data it has archived, a server will never archive data that it has already archived. When an archive is performed, the server only archives the data that has not yet been archived by that server (archiving is incremental).
You can view information about the data that has been archived by a particular server by using the Archive Volume Manager (right-click on the Server icon in the Windows task bar then select the Archive Volumes option from the context sensitive menu). The Archive Volume Manager displays a list of all the data that has been archived, and the list is organized according to the volumes (disks) on which the data is stored. You can also use it to create replacement volumes.
Types of Archive Media
ClearSCADA can archive data and events to any type of disk supported by Windows, including CDs, DVDs, and hard disks. When archiving to CD or DVD, Third-Party packet writing software is required. Unfortunately, Third Party packet writing applications can be unreliable and can even cause data loss. For this reason, we recommend that you archive data and events to hard disk, DVD-RAM or any other disk based archiving technology as these are more robust and reliable than archiving directly to CD or DVD.
ClearSCADA will archive data and event messages to your chosen archive media whenever a user selects the Archive pick action or the Archive method is called by a Schedule, Logic program etc.
When the archive media (hard disk, DVD-RAM etc.) becomes full, ClearSCADA raises an alarm and cannot complete the current archive task. In this situation, you need to replace the full disk with a blank disk and perform another archive so that the remaining data is archived to the new disk.
Recommendations when using Archiving
We recommend that you always replace each disk before it becomes full. This makes it easier to search for archived data at a later date---if all the data for a particular period is archived on the same disk, it is easier to locate the data you need.
You may wish to backup the archived data, for example, for secure offsite storage. You can do this by using Windows or a Third-Party application to copy the contents of an archive disk to another disk. In a multi-server system such as a Hot-Standby system, each server can independently perform archives, so there are at least two copies of the archived data (one per server). This is dependent on the server configuration, for more information, see ClearSCADA Help.
Note: On multi-server systems, the actual archiving process takes place on the server that is connected to the archive media, irrespective of the server status. This means that standby servers can archive data and remain in standby mode - all they require from the Main server is the instruction to initiate the archiving process. In return, the standby server will send archive status data to the Main server when the archive process is complete.
Most users choose to periodically archive data, for example, once a week. It is also common practice to replace each archive disk at regular intervals, for example, every 6 months.
We always recommend that you :
- Apply a unique label to the archive media (CD, DVD etc.) before the first archive takes place
- Make back-ups of all archive media
- Ensure all removable archive media is stored in a secure location.
Go: Home Back
Author
Link copied. Please paste this link to share this article on your social media post.
