INTEGRATION-ARCHITECTURE-DESIGNER practice question 8 of 25

Northern Trail Outfitters' (NTO) Salesforce org usually goes through 8k-10k batches a day to synch data from external sources. NTO's Integration…

Full set: 106 questions

Domain Design an Integration Solution

Northern Trail Outfitters' (NTO) Salesforce org usually goes through 8k-10k batches a day to synch data from external sources. NTO's Integration Architec has received requirements for a new custom object, FooBarc, for which 90M records will need to be loaded into the org. Once complete, 20GB (about 30M records) needs to be extracted to an external auditing system. What should the architect recommend using to meet these requirements in a day?
Answer options
A: Insert using Bulk API 2.0 and query using REST API.
B: Insert and query using Bulk API 1.0.
C: Insert using Bulk API 1.0 and query using REST API.
D: Insert and query using Bulk API 2.0. Correct answer
Correct Answer:
D. Insert and query using Bulk API 2.0.
Explanation

The requirement involves loading 90 million records and extracting 30 million records within a single day. This constitutes a Large Data Volume (LDV) scenario, for which the Bulk API is the appropriate tool. Bulk API 2.0 is the optimal choice for both operations. For the insert, it uses a record-based limit of 150 million records per 24 hours, which easily accommodates the 90 million record load. In contrast, Bulk API 1.0 has a batch limit of 15,000 per 24 hours. Given the existing 8k-10k batches, adding ~9,000 more batches for the new load would exceed this limit. For the extraction, Bulk API 2.0 is also designed for large-scale queries, simplifying the process by automatically handling data chunking.

Why Incorrect

A. Insert using Bulk API 2.0 and query using REST API. Using the REST API for a 30 million record query is incorrect. It's designed for small payloads and would be extremely inefficient, likely failing due to governor limits.
B. Insert and query using Bulk API 1.0. Inserting 90 million records with Bulk API 1.0 would create approximately 9,000 batches, which, when added to the existing 8k-10k daily batches, would exceed the 15,000 batch limit.
C. Insert using Bulk API 1.0 and query using REST API. This option combines the incorrect approaches from options A and B. The insert would exceed batch limits, and the query is not feasible with the REST API for this data volume.

References

1. Salesforce Developer Documentation

"Bulk API 2.0 and Bulk API Developer Guide":

Introduction: "Bulk API 2.0 provides a simple interface to quickly load large amounts of data into your Salesforce org and to perform bulk queries on your org data... You don’t have to break the data into batches yourself; we do the batching for you." This supports using Bulk API 2.0 for both large inserts and queries.

"Which API Should I Use?" Section: This guide explicitly recommends Bulk API for loading more than 50

000 records and states that REST API is suitable for real-time

small-batch operations

making it inappropriate for the 30M record extract.

2. Salesforce Developer Documentation

"Bulk API and Bulk API 2.0 Limits":

Bulk API Limits Table: Specifies the "Number of batches per rolling 24-hour period" for Bulk API is 15

000. This confirms that option B would likely fail.

Bulk API 2.0 Limits Table: Specifies the "Number of records processed per rolling 24-hour period for Bulk API 2.0 ingest jobs" is 150

000

000. This confirms the 90M record load is well within the limits for Bulk API 2.0.

3. Salesforce Architects

"Architect's Guide to Large Data Volumes":

Chapter: Data Loading Considerations: This guide emphasizes using the Bulk API for LDV scenarios. It highlights that Bulk API 2.0 simplifies the process by removing the need for manual batch management

which is a key advantage in this scenario where Bulk API 1.0 batch limits would be a problem.

Scroll to Top