|
Key Takeaways
|
|
While most automation tools focus on workflows, integrations, and real-time monitoring, they cannot often derive meaningful insights from operational data made available by automated workflows. This hinders data-driven decision-making. Automation Analytics & Intelligence (AAI) fills this critical gap by delivering actionable Data Insights, empowering SLA managers, business service owners, and SRE leaders to make informed decisions and optimize automation strategies.
Our AAI product team has been hard at work rethinking and improving the user experience to help users more quickly gain meaningful insights needed to optimize workloads and identify problems before they impact mission-critical business processes. As part of the user experience modernization initiative, the team has completely redesigned the experience with dynamic and interactive Dashboards and Data Insights. The first Data Insights and new Dashboards are available in the AAI 6.5.3 release.
The following is an interview with Jennifer Chisik, Head of Product for AAI. It has been edited for clarity.
Data Insights is a brand new feature area for us available with the AAI 6.5.3 release. Data Insights are organized, complex data extractions based on templates providing real-time views of your workload automation landscape. These next-generation Data Insights are dynamic, interactive, and shareable with other users and stakeholders. You can print, email, and schedule all Data Insights. Data Insights can be added to a dashboard as widgets that can be mixed and matched to create a customized dashboard that doesn’t lose any Data Insights dynamic qualities.
Data Insights can be dynamically filtered and sorted with various drill-downs. They can be scheduled and/or delivered via email directly from the UI to inform line of business stakeholders or executives who may not be using AAI day to day.
AAI creates a data warehouse that accumulates large volumes of consolidated, historical workload automation data from multiple scheduling vendors and platforms. That data warehouse and proprietary algorithms become a resource to produce useful derivative data such as trends, predictive analytics, SLA management metrics, and more. AAI collects data such as job conditions, calendar data, time conditions, and more. AAI then combines the data with our own proprietary statistical analysis, runs it through a proprietary modeling engine, and produces accurate and detailed predictions. This is what makes AAI so powerful.
Generic monitoring and observability tools, not specialized for automation, are limited by the underlying data they have access to. They do not track dependencies, which limits their ability to derive predictions. They don’t operate on a data warehouse containing rich historical data and are limited to live data available in the system, which reduces accuracy.
The first set of Data Insights, released with AAI 6.5.3, offers valuable insights into many of these data dimensions, and future releases will incorporate many more.
With the release of AAI 6.5.3, we've added Audit History, Processing Load, and Job Status Data Insights, and more are on the way.
The Audit History Data Insight type surfaces information about events and object changes within the underlying automation engine. This allows the user to understand who made what changes to which automation objects and when those changes were made. This Data Insight in AAI makes it easy to view and analyze this data, which is critical for audit compliance purposes and very difficult to get directly from the underlying automation engine. This feature is currently available specifically for AutoSys.
The Processing Load Data Insight type provides a clear picture of scheduler performance by showing the number of jobs running during a time interval and the total time in execution delays for each scheduler or machine during each slice of time. The graph, which can zoom in and out, reveals spikes and valleys of processing load. Because you can create a Processing Load Data Insight for totals by machine, it becomes a tool to support charge-back models for application teams.
The Job Status Data Insight type gives a quick overview of all the jobs in AAI grouped according to their current run status. You can create several Job Status Data Insights, filtering each to focus on a different run status or scheduler. This allows the user to create customized dashboards to focus on target areas of their automation processing. For example, a Job Status Data Insight can be created to show all job failures within a specific application, removing extraneous noise and allowing the user to focus on critical risks.
We released a new dashboard framework with AAI 6.5.3 that presents widgets based on defined Data Insights onto one page. Each widget is a window to the defined Data Insight. You can size the widgets to provide a miniature view of the Data Insight or expand them to fill the width of the page. This structure enables the creation of personalized overviews of the workload landscape, serving the varied user needs of different categories of users such as SRE leads, SLA managers, and business service owners.
You can act on widgets just as you would on the source Data Insight, including filtering, sorting, printing, and editing the definition. This makes your dashboard not just an overview but a valuable tool for monitoring and researching your workloads.
We plan to release new Data Insights quarterly. Trending and job run history are coming soon. In addition, we are planning some exciting new business user dashboards that will be surfaced as Data Insights.
We are always interested in customer feedback to inform our future direction. There are several ways to do this, including the AAI Community, through Broadcom Support, and by contacting me directly at jennifer.chisik@broadcom.com.
For more on what’s available in the latest AAI release, read the 6.5.3 Release Notes and watch the Data Insights demo.
Visit AAI on Broadcom Software Academy to learn more about AAI capabilities through the blog and other content, short demos, and free training.