---
title: "Short note: DW process architecture"
subject: "Data Mining"
chapter: 1
key: "semester_7/data_mining:dw-process-architecture-note"
source: https://www.bctengineer.com/q/semester_7/data_mining/dw-process-architecture-note
site: BCT Engineering
---

# Short note: DW process architecture

Data Mining · Chapter 1: Introduction

- Asked in: year not given, 3 marks
- Video: [Introduction to Data Warehouse (Data Architecture | Data Warehouse)](https://www.youtube.com/watch?v=vv0ReKrEQf4) — Software Architecture Academy
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_7/data_mining/dw-process-architecture-note
- Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-1

## Question

Write a short note on DW process architecture.

## Short answer

The data warehouse process architecture consists of three main managers: the load manager extracts and loads source data, the warehouse manager cleans, transforms, and aggregates data, and the query manager handles user queries. These processes work together with storage areas for detailed, summary, and archival data, plus metadata.

## Answer

The **data warehouse process architecture** describes the main processes (managers) that run in a warehouse and how data flows between them.

```text
 Sources -> LOAD MANAGER -> WAREHOUSE MANAGER -> Storage
                                                   |
                          Users <- QUERY MANAGER <-+
```

- **Load manager (front end)**: extracts data from sources, does simple transformations and loads it to a staging area or warehouse.
- **Warehouse manager**: runs inside the warehouse; cleans, transforms and merges data, builds indexes, views and aggregations, performs backup and archiving, and maintains metadata.
- **Query manager (back end)**: handles user queries, routes them to detailed or summary data, controls access, and schedules queries.
- **Storage areas**: detailed data, lightly and highly summarised data, archive, and metadata.

Each process is a separate piece of software (or team) with defined responsibilities, which makes the system easier to build, manage and scale.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. What is the data flow order?
   A. Users, storage, sources
   B. Query manager, warehouse manager, load manager
   C. Warehouse manager, load manager, users
   D. Load manager, warehouse manager, storage, query manager to users
   Answer: D. Sources feed the load manager, then the warehouse manager and storage; the query manager serves users. (Key point: Process order)

2. Which is done by the warehouse manager?
   A. Controlling user access
   B. Extracting from sources
   C. Cleaning, merging, building indexes/views/aggregations, backup and metadata
   D. Routing queries
   Answer: C. It manages data within the warehouse. (Key point: Warehouse manager)

3. Why separate the processes?
   A. It avoids using metadata
   B. It makes the system harder to manage
   C. Defined responsibilities make it easier to build, manage and scale
   D. It removes the need for storage
   Answer: C. Each process has clear responsibility, easing build and scaling. (Key point: Benefit of separate processes)

## Related in Data Mining

- [Metadata and DW architecture](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Ametadata-dw-architecture) — Data Mining, chapter 1
- [National data warehouse process architecture](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Anepal-national-data-warehouse) — Data Mining, chapter 1
- [Kathmandu Mall warehouse architecture](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Akathmandu-mall-warehouse) — Data Mining, chapter 1

## Related in other subjects

- [VHDL entity, architecture; D flip-flop](https://www.bctengineer.com/question.md?key=semester_6%2Fembedded_system%3Avhdl-entity-architecture-dff) — Embedded System, chapter 9