Importance of MetricBase

Importance of MetricBase

MetricBase is included in Now platform in Jakarta.This feature allows ServiceNow customers to collect, retain, analyze and visualize measurable data from any source or combination of sources on the Now Platform. It consumes and processes vast volumes of time series data and enables organizations to accelerate business decisions and actions with both historical perspective and predictive analytics.

So, What is Time-Series Data.

It is a collection of data points obtained sequentially at successive points in time. E.g.

  • Changes in memory and storage utilization on a server throughout the day

  • Minute to minute changes in office temperature recorded by a thermostat

  • Changes to blade RPM recorded by a monitoring device on a wind turbine

  • And many more

So, Why is MetricBase important?

Now-a-days, IOT devices are being used in vast areas. Organizations are challenged with an enormous volume of data metrics being collected continuously across an exploding number. This creates huge volumes of time series data.

Traditional relational databases and instance-based storage cannot accommodate the scale, performance and high-availability requirements for storing, aggregating and analyzing time series data.

MetricBase allows customers to store this data within the Now Platform and combine it with the data already available in the platform to accelerate decision-making and action.

MetricBase Architecture

MetricBase captures time series data in separate database accessed by the ServiceNow instance. It has been specifically designed for high capacity throughput, and can achieve capture rates of up to 160k values per second or 14 billion data points per day.


Data is pushed from external sources into MetricBase using the MetricBase REST API.

Once ingested into MetricBase, any Triggers configured for the Metric may take immediate action if a static threshold is breached. This may involve creating a case for someone to investigate, or initiating a Workflow that uses Orchestration to self-correct an issue.

Near real-time data can be also be presented in dashboards and reports, which allows clients to spot abnormal behavior and monitor trends over a period of time.

This data can also be analyzed through the use of Javascript or REST APIs that extract time series data over a specified period from MetricBase, and run Transforms (such as finding the average, maximum or minimum) on these data points

Storage Frequency and Retention Policies

Within MetricBase the most recent data is stored with higher fidelity, whilst older data is periodically compacted and stored with lower fidelity. MetricBase provides pre-defined retention policies that define the data collection interval for a metric.

Four tiers of Retention Policies are offered to clients. The table below outlines the number of days that data is retained at different fidelity levels for the provided policies.

Policy1 min10 mins30 mins1 hour2 hours1 day
 Coarse8 days8 days31 days397 days
 Medium31 days397 days
 High8 days31 days397 days
 Dense8 days31 days397 days

As an example, the Dense Retention Policy means:

  • For the last 8 days, all data points available at 1 minute granularity

  • For data points which are between 9 and 31 days old, we compress into 10 minute intervals

  • For data points older than 32 days, we compress into 1 hour intervals

Positioning Performance Analytics vs MetricBase

The table below compares the characteristics of Performance Analytics Indicators and MetricBase to provide some guidance on which option to choose.

Performance Analytics IndicatorsMetricBase
 Type of DataTask-oriented or human-generatedCMDB-oriented metrics, machine-generated data
 FidelityDailyPer Minute
 VolumeMillions of metrics

No comments

Leave a reply

Your email is never published nor shared. Required fields are marked *

Are you Human? * Time limit is exhausted. Please reload CAPTCHA.

Pin It on Pinterest

Share This