Scalability | Gimmal Docs

Records Management Web
What comprises the solution...	ASP.NET-based web application Data stored in SQL Provides a user interface for user interaction Provides a web API for external component integration such as connectors and client software integrations
How Scaling Works...	The web can be installed to multiple servers that all point to the same SQL server instance Network or hardware load balancing can be used to distribute among the multiple servers Session affinity should be set to "none"
When to Scale...	When CPU utilization is consistently above 90% for an extended duration more cores should be added or new servers should be added to the load balancer When Memory Pressure is consistently above 80% for an extended duration, more memory should be added or new servers should be added to the load balancer
General Sizing Guidelines...	(total User CountAvg. % Hourly Users)5)/60 = # of request per second Ceiling (# of requests per second/50) = # of servers Minimum of 2 servers recommended

Records Management Processing Service
What comprises the solution...	.NET-based Windows Service Data stored in SQL Server Performs reclassification of items Manages lifecycle for each item being tracked
How Scaling Works...	The service can be installed to multiple servers that all point to the same SQL Server instance The service supports failover, not load distribution The service acts as a controller off-loading the bulk of its workload to SQL Server stored procedures
Where to Scale...	Not Applicable
General Sizing Guidelines	Should have at least 2 servers for failover Recommended installing services on each Manager Web Server