Recommendations for Optimal Instance Performance

[GTSPerformance]

Performance is a matter that we can all agree on. We are sure customers and admins alike are not ecstatic when they receive notification of a default semaphore exhausted, instance response time high, or abnormal SQL response time alerts from Technical Support. In a perfect world, we would all want every single customer production environment to perform up to potential all the time. Unfortunately, this expectation breaks down in the face of reality. The truth is that every instance will deal with performance issues from time to time. The good news is that there are tools, methods, and information at the disposal of every single person desiring to improve their instance.

This article and the linked articles below are written by the Global Technical Support Performance team. We are a cross functional group of ServiceNow employees in the Technical Support department who live and breath the performance world. Many of ServiceNow's largest customers have benefited from the lessons presented in the documents below and we believe these lessons should be in the hands of every single person implementing ServiceNow.

Instance Performance Maintenance	Data Maintenance & UI Design	Server Side Scripting	Building Efficient Queries	IntegrationPerformance
Performance Maintenance and Administration How can I see current performance stats for ServiceNow?	Maintaining Task table Growth Demystifying table rotation, extension, and table cleaner Optimizing the Service Portal Typeahead search widget Reducing the burden of home pages  How can the Platform rebuild tables to reduce free space within them?  Performance Best Practice for Next Experience	Performance Best Practices for Server-side Coding in ServiceNow Performance Best Practice for Before Query Business Rules Caching data to improve performance Performance Best Practices for Coding Comma Separated Lists	Best Practice for Building Efficient Queries Improving Slow OR and JOIN Queries	Inbound Integration Troubleshooting (KB0564204)  Outbound Integrations Using SOAP / REST: Performance Best Practices An Introduction To The Batch REST API (/api/now/batch) Controlling session utilization from inbound integrations (sticky sessions)

Lets ensure all implementers of ServiceNow are equipped to build with performance in mind by bookmarking this article (and/or the linked KB) and sharing it with everyone who cares about performance; which should be anyone working on a ServiceNow Instance. The content in this KB will be updated as new information and methods are provided. We hope you will adopt the guidance in these articles as you establish your own quality control standards including Peer Review Checklists and the like.

This article is maintained by the ServiceNow Performance team in Global Technical Support. We are a cross-functional group of experts working at ServiceNow in the Support department. We are located around the world but share a passion to help our customers solve their performance issues! If you have questions about the content of this article or the linked articles, you can ask a question directly on the community by adding a “comment” and we will try to respond. However, if you have urgent questions or specific issues, please see the list of resources in the Bio section of our profile page: ServiceNowPerformanceGTS

Best regards,

Your Performance Team

Mirror article on ServiceNow Support Portal: KB0952746 - Performance Recommendations for Best Instance Management

[Michael83]

What a great series team! I'm glad I stumbled upon this. 

[German Alvarez1]

The link to

• Viewing statistics (xmlstats.do & stats.do) for all my nodes?

Does not work anymore!

Could you please support on recovering that content. Sounds promising!

[GTSPerformance]

@German Alvarez1 Thanks for the feedback! We've removed that article from our list since it was out of date and replaced it with the following one:

• How can I see current performance stats for ServiceNow?

We hope this is helpful!

Best regards,

Your ServiceNow Technical Support Performance team

[Anton42]

So, it is still (or again) not possible for customers to see the important actions of all nodes at once being able to debug in a meaningful way?

[GTSPerformance]

@Anton42  Thanks for your question and we apologize since it sounds like you are experiencing frustration!

1. Have you used some of the tools that have been available in the past - specifically, Application Insights and ServiceNow Performance Homepage? If so, what did you find lacking about those tools so that we can provide feedback to the development teams in charge of creating such tools so they can address those concerns in future tools?

2. Have you looked at the Impact Instance Observer (IO) tool? This is most powerful tool to see important details about all nodes going forward. We understand that not all customers will have access to this tool since it currently requires a paid Impact package. What specific feedback can we provide to that team?

In the articles linked above, we have endeavored to explain many of the most helpful tools for customers to understand performance of their instance and believe that there are meaningful tools to debug ServiceNow, but we are always working to improve customer's experience. Our team has also created a training course for customers to understand and diagnose performance issues in their environments that you can take here: Performance Troubleshooting Overview.

You can also send us a direct message with your feedback if you'd like. Thanks for helping us improve ServiceNow!

Best regards,

Your ServiceNow Technical Support Performance team