Salesforce Certified Data Architect 1-10

PK Chunking is a feature provided by the Bulk API (Bulk API 2.0 is now the mainstream), and it is a mechanism that splits large-scale data queries into multiple smaller batches by primary key (Id) range for processing.
This avoids query timeouts and improves processing performance.
Because processing large-scale data such as 15 million records in a single query is difficult, leveraging this feature is an effective measure.
The Tooling API, Metadata API, and Streaming API serve different purposes and are not suited to solving this problem.
PK Chunking

In master data management, a design that ensures data consistency, reliability, and history management is important.
First, it is necessary to clearly determine which data will be replicated and shared across systems (identifying the replication scope).
In addition, by defining the system of record that serves as the authoritative source for each set of data, duplication and inconsistency can be prevented.
Moreover, because long-term data must be handled, an archiving strategy is also indispensable.
On the other hand, a data warehouse and BI tools are for analytical purposes and are not essential elements of an MDM strategy.
Master Data Management

The key point of this question is addressing report timeouts caused by large data volumes.
The most effective approach is to aggregate and retain the data in advance, reducing the amount of processing at report execution time.
If a pre-aggregated custom object is prepared, there is no longer a need to process large volumes of raw data at report execution time, and performance improves dramatically.
The other options increase data redundancy, introduce external dependencies, or mismatch the intended purpose, so they do not provide a fundamental solution.
Reporting

This question assumes a mechanism that can track the handling status and changes of cases.
Therefore, the first step is to enable the capability that records the history of changes such as case status and owner reassignment.
By enabling field history tracking, you can analyze chronologically how a case progressed and when it was closed.
This makes it possible to build reports that help calculate accurate close rates and improve the handling process.
The other options are supplementary means, and acquiring the foundational data must come first.
Field History Tracking

This question concerns an archiving design that assumes a large data volume of 5 million records per year.
In Salesforce, leveraging Big Objects, which can retain large volumes of data without impacting performance, is effective.
In addition, using an external system (Heroku plus external objects) allows data to be retained outside Salesforce while remaining accessible on demand.
This avoids storage limits and performance degradation.
Custom objects have data volume constraints and are unsuitable, and a simple external DB integration alone provides insufficient integration with the Salesforce UI.
Big Objects

In large data migrations, querying parent record IDs in advance is inefficient and causes timeouts.
Therefore, it is important to resolve the parent-child relationships using external IDs and eliminate the query processing itself.
External IDs are indexed, and because upsert operations can efficiently associate parent and child records, high performance can be maintained even in large-scale migrations.
This eliminates the need to acquire IDs beforehand and greatly reduces processing time and load.
Apex Developer Guide

To go beyond the limits of standard reports, introducing an advanced analytics platform is necessary.
First, Einstein Analytics (now CRM Analytics) can integrate multiple data sources and deliver advanced visualization and AI analysis, providing insights beyond standard capabilities.
In addition, by leveraging AppExchange apps, you can quickly extend external tool integrations and advanced analytical features.
Weekly snapshots and the Setup Audit Trail have limited purposes and are not direct solutions for enhancing analytics.
CRM Analytics

In Lightning Experience, the legacy Notes and Attachments are deprecated, and migration to Files and Enhanced Notes is required.
Therefore, it is essential to migrate existing notes and attachments to the new data model using a migration tool.
In addition, Files have enhanced sharing, search, and security capabilities and have become the standard file management method in Lightning. Simply adding a related list or manually re-registering does not allow the new features to be leveraged and does not meet the requirements.
Convert Attachments and Classic Notes

When exporting large data volumes, distributing the processing load is important.
First, by splitting the data into multiple batches, you can distribute the processing load and avoid timeouts and errors.
In addition, using PK Chunking splits the data by primary key and allows large-scale data to be processed efficiently.
This also optimizes parallel processing and improves performance.
Parallel mode alone can cause contention, and simply limiting the batch size is not a fundamental solution.
PK Chunking

When integrating data across multiple Salesforce orgs, a mechanism for real-time synchronization and cross-org sharing becomes important.
First, Change Data Capture (CDC) delivers record create, update, and delete events in real time and achieves data synchronization with other orgs.
In addition, Salesforce to Salesforce is a standard feature that can directly share and synchronize records between orgs, achieving data integration simply.
Salesforce Connect is primarily for referencing external data, and Federation Search is for search purposes, so they are not suitable for this requirement.
Access Data in Another Salesforce Org with the Cross-Org Adapter for Salesforce Connect