Cracking the Code: “Steve's Number of Zeros”for Estimating Integration Volumes!
Introduction
Welcome to another episode in our integration series at Steve Tech Arc! Today, we’ll delve into the crucial aspect of estimating integration volumes during the design phase of an integration project, focusing on the methodology I term as “Steve’s Number of Zeros”.
Why Estimation Matters
Understanding the volume of data to be transferred between systems is pivotal. It informs the scaling decisions, choice of mechanisms, and components scaling, especially when using an integration layer like MuleSoft. The goal is to derive key metrics like Transactions Per Second (TPS) starting from rough business guidelines.
The Estimation Process
Baseline Assumptions: Begin with baseline assumptions that the business can comprehend and commit to. If exact numerical values are elusive, use “Steve’s Number of Zeros” to gauge whether you are dealing with tens, hundreds, thousands, or millions of records.
Conversion to Industry Metrics: Convert these rough numbers into industry-standard units of measurement crucial for integration estimation, like TPS, records per second, and bytes per second. Understand the activity iteration, whether the integration activity is sustained or peaked over specific hours.
Practical Estimation Examples
I’ve shared a sample template that aids in these calculations, focusing on daily activities, transactions, records, and byte sizes. The activity window is also determined, whether it’s for eight, twelve, or fewer hours of activity.
Example Scenarios:
Business Activity Transfer: For transferring closed opportunities to an ERP in real-time, you can estimate the number of accounts, opportunities closed per month per account, records per opportunity, and data size associated with each record.
Truck Deliveries Notification: For a scenario where trucks send various notifications, estimate the number of trucks, deliveries per day, notifications per delivery, and data size per notification.
Appointment Notifications: For sending appointment notifications to customers, estimate the number of stores, appointments per day, notifications per appointment, and data size per notification.
Case Handling by Agents: For agents working on cases, estimate the number of agents, cases handled per hour, records updated per case, and data size per transaction.
Conclusion: Practical Estimation for Integration Projects
Estimating integration traffic early in the project is key. When working with Salesforce, ensure you won’t exceed limits and scale the appropriate components for a smooth load across the activity window. Understanding the details of the implementation, like queues and workers, is also crucial for the integration team.
These methods and assumptions provide a practical approach to translating abstract business requirements into concrete details for your integration team, serving as a baseline for accurate estimations.
Stay Tuned
Thank you for diving into the tech swimming pool.
Stay tuned for more insights and surprises ahead here on www.SteveTechArc.com and to the @SteveTechArc YouTube channel. Subscribe and enhance your understanding of Salesforce and how you can integrate it with other systems.
Helping change the world by sharing integration info with fellow Architects and those on their Architect Journey!
STA 1.5
