How should I archive my historical data?

Questions addressed:

Page 1:

• What is historical

data?

• What impact does

historical data

have on

performance?

Page 2:

• How should I

archive historical

data?

• Who can set up

archive rules and

can the rules be

customized?

• What happens if I

need to access

archived data?

What is historical data?

The longer an instance runs, the more likely it is to accumulate data that’s no longer relevant. For example, task records from

two years ago are typically less relevant than currently active tasks. You may not be able to delete this data because you

need it for auditing or for historical purposes.

What impact does historical data have on performance?

Old data may eventually cause performance issues by consuming system resources and slowing down queries and reports.

Archiving historical data will ensure you get the optimal performance and keep core operations also running smoothly.

Insider’s Tips:

• Archiving is mostly beneficial when the table is large and actively used. The best candidates include task (incident,

sc_task, problem, change_request, etc.).

• Emails that are attached to incidents will not be automatically archived with those incidents. There is a separate tool

called Email Retention that will allow you to archive and eventually destroy email messages that you no longer need or

when your email table is excessively large.

Archive

Incident

(Incident)

Active data

Inactive

data

2,000,000 rows

400,000 rows (20%)

1600,000 rows (80%)

Before Archive

Incident

(Incident)

Active data

After Archive

400,000 rows (100%)

Who can set up archive rules and can the rules be customized?

Administrator roles can set up archiving rules. Archive Rules can be set-up from Archive Rules

section. Only one archive rule can be set-up per table. Rules can be set-up for both OOB tables

and custom tables.

You can set archive rules to control how many records an archive rule processes per batch job,

how much sleep time is between each archive rule batch job run (in seconds), and the maximum

number of batch jobs to run w ithin the hour.

By default, archive rules are set for 100 records for each batch job, sleeps 1 second between batch

jobs, and runs 10 batch jobs in an archive run (every hour. You can customize these settings by

updating system properties, including changing the process not to run in peak hours.

With the Paris release, archiving is now a producer/consumer model that can run parallel on

multiple nodes. Records to be archiv ed are now recorded in a queue table in batches. Pre-Paris

archiver properties don’t work with the Paris release.

What happens if I need to view data that has been archived?

You can view archived data, change an archive schedule,

and restore archived data.

When you restore a record, the instance inserts it back into the

primary table and flags the record as having been restored in

the log. It’s important to note that archive rules will not archive

restored records. In order to archive a restored record you must

manually archive the record.

With the Paris release, there is an option to re-archive restored

records again after a period of time.

How should I archive historical data?

Archive historical data with Data Archiving, which is enabled by default in ServiceNow. Archiving is a scheduled process that runs every hour and executes all archive r ules

one by one to remove them from immediate access and free system resources. (Note: Archiving is not a solution to reduce your database size.)

ACTIVATE

Activate data archiving. The

Data Archiving plugin installs the

System Archiving application.

CREATE DESTROY RULE

Create a destroy rule to

delete archived records after

a specified amount of time.

SET ARCHIVE RULES

Set the rules that will control

your archive process.

SCHEDULE ARCHIVE DELETION

Configure the archive destroyer

batch process w hich runs the

archive destroy rules.

Contact your account team who can reach out to SWAT to

assist you with any archiving database issues.

like to be a contributor to future ServiceNow best practice

content.

Insider’s tip:

It’s better to use more batches than increase the size of the batch. This helps avoid replication lag.

How should I archive my historical data? (Continued)