Back to documents

Document 21: annotated-Viet_Anh_Doan_541031778_A1_Report.pdf

Status: ready

S3 bucket: a2-pdf-bucket-andy

S3 key: uploads/1779799429_0685b3c234304853adebd4850814552b_annotated-Viet_Anh_Doan_541031778_A1_Report.pdf

Uploaded: 2026-05-26 12:43:49.102316+00:00

Processing Runs

Strategy Status Chunks Average length Processing time Error
Fixed-size chunking completed 4 948.0 0.291 sec
Paragraph-aware chunking completed 3 1063.0 0.287 sec

Sample Chunks

Fixed-size chunking

Chunk 0 - 1000 characters

# Page 1

C O M P5 3 4 9-  A1  D E P L O Y M E N T  R E P O R T 
Student Name: Viet Anh Doan - SID: 541031778 
1. Architecture Diagram

# Page 2

2. Services and Interactions 
Service Description Role Instance 
type 
Amazon EC2 Amazon Computing Service, 
that provides virtual servers 
(instances) to deploy the 
application on AWS cloud. 
 
Provide an EC2 instance to host the 
Flask web application and handle 
all user requests 
t3.small 
 
Amazon S3 Amazon Simple Storage 
Service, that provides 
scalable and reliable storage 
option 
Store users’ pdf documents in S3 
bucket after user send upload 
request to backend server 
AWS 
Managed 
Service 
Amazon RDS 
(PostgreSQL) 
Fully managed service from 
AWS that simplifies setting 
up, operating, and scaling 
relational database. 
Store documents’ metadata and 
summary after users upload the 
document, and AI summary request 
is returned. 
d...

Chunk 1 - 1000 characters

nts’ metadata and 
summary after users upload the 
document, and AI summary request 
is returned. 
db.t3.micro 
 
Google Gemini 
API 
External APIs for Generative 
AI features 
Receives user’s pdf document, and 
return an AI-generated summary 
 
 
End-to-end flow 
The user uploads a PDF via a POST endpoint to doc -ec2, which saves it temporarily in /tmp, uploads it to 
S3 (amazon-s3-doc-bucket) and Gemini Files API. Gemini processes the file and returns an AI -generated 
summary, which doc-ec2 saves along with the document metadata to RDS (doc -db). The /tmp file is then 
cleared and the user is redirected to the result page displaying the summary.  
3. Key Configuration details 
Service Configuration 
EC2 to 
Internet 
Deployed in public subnet ( 172.31.0.0/20) in ap-southeast-2a. Security group opens port 22 
(SSH) and port 5000 (Flask). Internet Gateway enables public access.. 
EC2 to...

Paragraph-aware chunking

Chunk 0 - 144 characters

# Page 1

C O M P5 3 4 9-  A1  D E P L O Y M E N T  R E P O R T 
Student Name: Viet Anh Doan - SID: 541031778 
1. Architecture Diagram

# Page 2

Chunk 1 - 2635 characters

2. Services and Interactions 
Service Description Role Instance 
type 
Amazon EC2 Amazon Computing Service, 
that provides virtual servers 
(instances) to deploy the 
application on AWS cloud. 
 
Provide an EC2 instance to host the 
Flask web application and handle 
all user requests 
t3.small 
 
Amazon S3 Amazon Simple Storage 
Service, that provides 
scalable and reliable storage 
option 
Store users’ pdf documents in S3 
bucket after user send upload 
request to backend server 
AWS 
Managed 
Service 
Amazon RDS 
(PostgreSQL) 
Fully managed service from 
AWS that simplifies setting 
up, operating, and scaling 
relational database. 
Store documents’ metadata and 
summary after users upload the 
document, and AI summary request 
is returned. 
db.t3.micro 
 
Google Gemini 
API 
External APIs for Generative 
AI features 
Receives user’s pdf document, and 
return an AI-generated summary 
 
...

Query Comparison