Skip to content

Commit 920d44c

Browse files
pareenavermapareenaverma
authored
Merge pull request #3012 from odidev/qdrant_LP
Build Semantic Search and Chatbot Retrieval Systems with Qdrant on Google Cloud C4A Axion processors
2 parents 7a6d2bd + c3548b1 commit 920d44c

13 files changed

Lines changed: 1016 additions & 0 deletions

File tree

‎content/learning-paths/servers-and-cloud-computing/qdrant-on-axion/_index.md‎

Lines changed: 73 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,73 @@
1+
---
2+
title: Build Semantic Search and Chatbot Retrieval Systems with Qdrant on Google Cloud C4A Axion processors
3+
4+
draft: true
5+
cascade:
6+
draft: true
7+
8+
minutes_to_complete: 30
9+
10+
who_is_this_for: This is an introductory topic for developers, data engineers, and platform engineers who want to build semantic search systems and chatbot retrieval pipelines on Arm64-based Google Cloud C4A Axion processors using the Qdrant vector database.
11+
12+
learning_objectives:
13+
- Deploy and run the Qdrant vector database on Google Cloud C4A Axion processors
14+
- Generate vector embeddings using transformer models
15+
- Store and index embeddings efficiently using Qdrant
16+
- Perform semantic similarity search using vector queries
17+
- Build a simple chatbot retrieval system powered by vector search
18+
19+
prerequisites:
20+
- A [Google Cloud Platform (GCP)](https://cloud.google.com/free) account with billing enabled
21+
- Basic familiarity with Python
22+
- Basic understanding of machine learning embeddings
23+
- Familiarity with Linux command-line operations
24+
25+
author: Pareena Verma
26+
27+
##### Tags
28+
skilllevels: Introductory
29+
subjects: Databases
30+
cloud_service_providers:
31+
- Google Cloud
32+
33+
armips:
34+
- Neoverse
35+
36+
tools_software_languages:
37+
- Qdrant
38+
- Python
39+
- Sentence Transformers
40+
- Docker
41+
42+
operatingsystems:
43+
- Linux
44+
45+
# ================================================================================
46+
# FIXED, DO NOT MODIFY
47+
# ================================================================================
48+
49+
further_reading:
50+
- resource:
51+
title: Google Cloud documentation
52+
link: https://cloud.google.com/docs
53+
type: documentation
54+
55+
- resource:
56+
title: Qdrant documentation
57+
link: https://qdrant.tech/documentation/
58+
type: documentation
59+
60+
- resource:
61+
title: Sentence Transformers documentation
62+
link: https://www.sbert.net/
63+
type: documentation
64+
65+
- resource:
66+
title: Vector Databases Explained
67+
link: https://qdrant.tech/articles/what-is-a-vector-database/
68+
type: documentation
69+
70+
weight: 1
71+
layout: "learningpathall"
72+
learning_path_main_page: yes
73+
---

‎content/learning-paths/servers-and-cloud-computing/qdrant-on-axion/_next-steps.md‎

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,8 @@
1+
---
2+
# ================================================================================
3+
# FIXED, DO NOT MODIFY THIS FILE
4+
# ================================================================================
5+
weight: 21 # Set to always be larger than the content in this path to be at the end of the navigation.
6+
title: "Next Steps" # Always the same, html page title.
7+
layout: "learningpathall" # All files under learning paths have this same wrapper for Hugo processing.
8+
---

‎content/learning-paths/servers-and-cloud-computing/qdrant-on-axion/architecture.md‎

Lines changed: 120 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,120 @@
1+
---
2+
title: Architecture
3+
weight: 8
4+
5+
### FIXED, DO NOT MODIFY
6+
layout: learningpathall
7+
---
8+
9+
# Chatbot Architecture Using Qdrant
10+
11+
In this section, you explore the architecture behind the chatbot retrieval system built using Qdrant on Google Axion Arm-based infrastructure.
12+
13+
This architecture demonstrates how modern AI systems perform **semantic similarity search** to retrieve relevant information from stored knowledge.
14+
15+
Unlike traditional keyword search systems, vector databases allow applications to retrieve information based on **semantic meaning and contextual similarity**.
16+
17+
18+
## System architecture
19+
20+
The chatbot system retrieves relevant information through vector embeddings and similarity search.
21+
22+
```text
23+
User Question
24+
│
25+
▼
26+
Embedding Model
27+
(Sentence Transformer)
28+
│
29+
▼
30+
Vector Representation
31+
│
32+
▼
33+
Qdrant Vector Database
34+
(Vector Similarity Search)
35+
│
36+
▼
37+
Top Matching Knowledge
38+
│
39+
▼
40+
Chatbot Response
41+
```
42+
43+
44+
45+
## Components
46+
47+
**Embedding Model**
48+
49+
The embedding model converts text into numerical vectors representing semantic meaning.
50+
51+
**Example model used:**
52+
53+
```text
54+
sentence-transformers/all-MiniLM-L6-v2
55+
```
56+
57+
This lightweight transformer model is commonly used for semantic search and AI retrieval workloads.
58+
59+
## Vector Database (Qdrant)
60+
Qdrant stores and indexes vector embeddings generated from documents and user queries.
61+
62+
It enables fast **nearest-neighbor similarity search**, which finds the most relevant vectors based on semantic similarity.
63+
64+
Key capabilities:
65+
66+
- high performance vector indexing
67+
- semantic similarity search
68+
- scalable vector storage
69+
70+
## Knowledge Base
71+
72+
The system stores knowledge documents such as:
73+
74+
- technical documentation
75+
- support articles
76+
- FAQs
77+
- internal company knowledge
78+
79+
During ingestion, these documents are converted into embeddings and stored in Qdrant.
80+
81+
## Chatbot Query Engine
82+
83+
When the user asks a question:
84+
85+
1. The query is converted into an embedding
86+
2. Qdrant searches for the closest vectors
87+
3. The chatbot returns relevant information
88+
89+
This process enables the chatbot to understand intent and meaning, rather than relying solely on keyword matching.
90+
91+
## Benefits of This Architecture
92+
93+
This design provides several advantages:
94+
95+
- semantic search instead of keyword matching
96+
- scalable knowledge retrieval
97+
- faster query responses
98+
- efficient AI workloads on Arm infrastructure
99+
100+
## Running on Axion
101+
102+
This example demonstrates that Axion Arm infrastructure can efficiently run vector search workloads.
103+
104+
- Benefits include:
105+
- energy-efficient compute
106+
- scalable cloud infrastructure
107+
- optimized performance for AI workloads
108+
109+
## What you've learned
110+
111+
In this section, you learned how the chatbot retrieval system works using vector search.
112+
113+
You explored:
114+
115+
- How embeddings represent semantic meaning
116+
- How Qdrant stores and indexes vectors
117+
- How similarity search retrieves relevant knowledge
118+
- How this architecture supports chatbot and RAG systems
119+
120+
Together, these components form the foundation for modern AI-powered search and knowledge retrieval systems running on Arm-based cloud infrastructure.

‎content/learning-paths/servers-and-cloud-computing/qdrant-on-axion/background.md‎

Lines changed: 47 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,47 @@
1+
---
2+
title: Get started with Qdrant on Google Axion C4A
3+
weight: 2
4+
5+
layout: "learningpathall"
6+
---
7+
8+
## Explore Axion C4A Arm instances in Google Cloud
9+
10+
Google Axion C4A is a family of Arm-based virtual machines built on Google’s custom Axion CPU, which is based on Arm Neoverse-V2 cores. Designed for high-performance and energy-efficient computing, these virtual machines offer strong performance for data-intensive and analytics workloads such as big data processing, in-memory analytics, columnar data processing, and high-throughput data services.
11+
12+
The C4A series provides a cost-effective alternative to x86 virtual machines while leveraging the scalability, SIMD acceleration, and memory bandwidth advantages of the Arm architecture in Google Cloud.
13+
14+
These characteristics make Axion C4A instances well-suited for modern analytics stacks that rely on columnar data formats and memory-efficient execution engines.
15+
16+
To learn more, see the Google blog [Introducing Google Axion Processors, our new Arm-based CPUs](https://cloud.google.com/blog/products/compute/introducing-googles-new-arm-based-cpu).
17+
18+
## Explore Qdrant Vector Search on Google Axion C4A (Arm Neoverse V2)
19+
20+
Qdrant is an open-source vector database designed for efficient similarity search and high-performance vector indexing. It enables applications to store and retrieve embeddings—numerical representations of data such as text, images, or audio—allowing systems to perform semantic search and AI-powered retrieval.
21+
22+
Vector databases like Qdrant are commonly used in modern AI systems to support applications such as semantic search, recommendation systems, anomaly detection, and Retrieval-Augmented Generation (RAG) pipelines. By storing embeddings and performing nearest-neighbor search, Qdrant allows applications to retrieve the most relevant information based on semantic meaning rather than simple keyword matching.
23+
24+
Running Qdrant on Google Axion C4A Arm-based infrastructure enables efficient execution of AI and vector search workloads. Axion processors, based on the Arm Neoverse V2 architecture, provide high performance and improved energy efficiency for modern cloud-native applications and data services.
25+
26+
Using Qdrant on Axion allows you to achieve:
27+
28+
- High-performance vector similarity search for AI applications
29+
- Efficient embedding, storage, and indexing for semantic retrieval
30+
- Low-latency data access for chatbots and AI assistants
31+
- Scalable infrastructure for Retrieval-Augmented Generation (RAG) pipelines
32+
- Cost-efficient execution of vector database workloads on Arm-based cloud infrastructure
33+
34+
Common use cases include AI chatbots, semantic search engines, recommendation systems, enterprise knowledge assistants, document retrieval systems, and machine learning feature stores.
35+
36+
To learn more, visit the [Qdrant documentation](https://qdrant.tech/documentation/) and explore how vector databases enable modern AI applications.
37+
38+
## What you've learned and what's next
39+
40+
In this section, you learned about:
41+
42+
* Google Axion C4A Arm-based VMs and their performance characteristics
43+
* Qdrant as a vector database for storing and retrieving embeddings
44+
* Semantic similarity search and how it powers AI retrieval systems
45+
* How vector search enables chatbot and RAG-style knowledge retrieval
46+
47+
Next, you can explore how to extend this setup by integrating large language models (LLMs) to build a full Retrieval-Augmented Generation (RAG) pipeline, enabling AI systems to generate context-aware responses using information retrieved from the Qdrant vector database.

0 commit comments

Comments
 (0)