Scaling integrations with bulk, async and big objects
What makes an integration scale: Bulk API 2.0 versus bulkified REST by volume and latency, cutting API calls with composite requests, automatic PK chunking for big extracts, Batch Apex concurrency and callout sizing, bulkified triggers and the shared async limit, 70% headroom with production-scale load tests, big objects for long-term logs and event fan-out for new consumers.
Given a use case, describe what should be considered when building a scalable solution.
~8 min read · 11 practice questions ·
What the exam expects
- Choose Bulk API 2.0, bulkified synchronous REST or virtualization from volume, latency and where the data must live.
- Reduce API consumption with composite requests and bulk jobs, and extract very large tables without timeouts.
- Size Batch Apex for concurrency and callouts, bulkify triggers on loaded objects, and avoid flooding the shared async limit.
- Set scalability standards: headroom below governor limits, production-scale load tests, big objects for multi-year logs, events for growing consumer lists.
Key terms: bulkified (one operation handles a collection, not one record); PK chunking (splitting a bulk query by record ID ranges); flex queue (where submitted batch jobs wait in Holding status); headroom (the share of a limit left unused for spikes); big object (a horizontally scalable store for hundreds of millions of records).
Key facts
- Detail
- More than 2,000 records; fewer use bulkified REST or SOAP
- Detail
- Whole request counts as one API call
- Detail
- Org-wide aggregate per 24 hours, not per user
- Detail
- Automatic in Bulk API 2.0 queries; manual header in original Bulk API
- Detail
- 5 queued or active; 100 Holding in the flex queue
- Detail
Database.AllowsCallouts; up to 100 per start, execute or finish
- Detail
- 250,000 or licenses x 200, whichever is greater; org-wide
- Detail
- 70% of governor limits under normal load
- Detail
- Consistent performance up to billions of records
Sources for this section (9)
- Bulk API introdeveloper.salesforce.com (opens in a new tab)
- Compositedeveloper.salesforce.com (opens in a new tab)
- sObject Collectionsdeveloper.salesforce.com (opens in a new tab)
- API Request Limitsdeveloper.salesforce.com (opens in a new tab)
- Bulk API 2.0 vs Bulk APIdeveloper.salesforce.com (opens in a new tab)
- Use Batch Apexdeveloper.salesforce.com (opens in a new tab)
- Execution Governors and Limitsdeveloper.salesforce.com (opens in a new tab)
- Reliability (Well-Architected)architect.salesforce.com (opens in a new tab)
- Big Objectsdeveloper.salesforce.com (opens in a new tab)
- Why does a Bulk API 2.0 extract not need the PK chunking header?
- When should a load use bulkified REST instead of Bulk API 2.0?
- How many batch jobs can be queued or active at once in an org?
- Why load test in a Full Copy sandbox instead of a small sample?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Integration Architect pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 25 lessons of the Integration Architect guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Inline practice questions and one-click topic practice (11 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Unlimited timed practice exams