Amazon DynamoDB

TCSS562: Group 7
Resham Ahluwalia
Savita Rana
SriHarshitha Somaraju
Priyanka Konduru
1
Presentation Overview
The key points of the presentation:

● Introduction and History of DynamoDb

● Amazon released DynamoDb on January 18, 2012.

● Amazon DynamoDB lets you create a database table that can store and
retrieve any amount of data, and serve any level of request traffic.

● Supports both document and key-value store models.

3
Motivation
The reason for the introduction of NoSQL is because RDBMS cannot actually cope
with some modern applications.This is because RDBMS has the following
problems:
● Low throughput
● Only vertical scaling
● Object –Relational mapping is expensive
● The thinking "one size fits all" is flawed.
● ACID (Atomicity, Consistency, Isolation and Durability) is not always needed.

4
Features
● Automatic data replication
● Infinitely scalable read-write I/O running on IOPS-optimized solid state drives.
● A provisioned-throughput model
● Amazon DynamoDB Accelerator (DAX)
● Data backed up to S3
● Integrated with other AWS services
● Pay-per-use model
● Security and access control

5
Use Case - I
Redfin Manages Data on Hundreds of Millions of Properties Using Amazon
DynamoDB

6
Use Case - II
Expedia’s real-time analytics applications uses DynamoDB

7
Use Case - III
Nexon uses DynamoDb as its primary game database for a new blockbuster
mobile game, HIT

8
Advantages
Fast, Consistent Performance Highly Scalable Fully Managed

Event Driven Fine-grained

9
Disadvantages
● 64 KB limit on row size
● Consistency comes with cost
● Attributes values cannot be null or empty
● Joins are impossible
● Indexing

10
Usability
● Supports various platforms (.NET, PHP, Java, Python, Ruby, etc)
● The pricing is very simple
● The data model is very flexible, use of JSON
● Auto Scaling
● DAX, in-memory cache that can reduce DynamoDB response times from
milliseconds to microseconds making its usability very easy
● Fully managed cloud database

11
DynamoDb Data
Model

● DynamoDb is
schemaless,except primary
key
● Each item can have any
number of attributes.
● An attribute is a name-value
pair
12
Data model
concepts
● Primary keys
○ Hash type
○ Hash and Range type
● Secondary Indexes
○ Local Secondary
○ Global Secondary
● DynamoDb data types
○ Scalar Data types
○ Multivalued Data types

13
Cost Discussion
● Terms involved in DynamoDB pricing:
○ WCU: Number of item writes per second x item size in 1KB blocks
○ RCU: Number of item reads per second x item size in 4KB blocks
○ Target utilization= Real consumed capacity / total provisioned capacity

Pricing:

Resource Type Details Monthly Price

Provisioned Throughput (Write) One write capacity unit (WCU) provides up to one write per second, enough for As low as $0.47 per WCU
2.5 million writes per month

Provisioned Throughput (Read) One read capacity unit (RCU) provides up to two reads per second, enough for As low as $0.09 per RCU
5.2 million reads per month

Indexed Data Storage DynamoDB charges an hourly rate per GB of disk space that your table As low as $0.25 per GB
consumes

14
Cost Discussion
AutoScaling Example:
Create a new table in US East (N.Virginia) region
Target utilization-70%
Minimum capacity units to 100 RCUs and 100 WCUs
Maximum capacity set to default limit of 40,000 RCUs and 40,000 WCUs.

During first hour: AutoScaling will not be triggered if actual utilization is with in target utilization
Bill for the hour would be $0.065 for the 100 WCUs provisioned ($0.00065 * 100), and $0.013 for the 100 RCUs
($0.00013 * 100).

During second hour:If consumed utilization increased to 100WCU’s and 100RCU’s, resulting in provisioned
capacity of 143 RCUs and 143 RCUs (% of 100 consumed ÷ 143 provisioned).
The bill for this second hour would be $0.0925 for 143 WCUs and $0.01859 for 143 RCUs.

Monthly bill for the table is the sum total of the hourly pricing of the capacity provisioned by Auto Scaling through
the month.

15
Cost Example(US East Region):
Number of requests requests per second -100 (Writes)
Number of reads per second - 200 (This data is read twice)
Data Size -2Kb
Cost for S3( Monthly):
Total Number of Write Requests : 30*24*60*60*100= 259.2 Million Requests per month
Total Storage- 494.3GB
Cost for write requests - $1296 (Pricing: PUT, COPY, POST, or LIST Requests-$0.005 per
1,000 requests )
Cost for reads-$207.36 (Pricing:$0.0004 per 1,000 requests)
Cost for storage: $11 (0.023 per GB)
Cost for data returned: $0.346 (Data Returned by S3 Select-$0.0007 per GB)
Total Cost: $1500 Approximately

16
Cost Using DynamoDB
Number of WCUs- 200(Since 2KB)
Monthly Cost- 200WCUs * $0.47per WCU= $94
Number of RCUs- 200 (Since data is read twice)
Monthly Cost- 200RCUs * .09per RCU= $18
Cost for storage-494.3GB * .25per GB= $123.575
Total Cost- $235(6X times cheaper than S3)
That’s not all.
If data size per request is 50KB instead of 2KB.
Cost for S3 would be approximately $3000 per month whereas DynamoDB would cost
approximately 6500 (Almost twice expensive as S3)

17
Alternatives
Features DynamoDB Google Big Table

Architecture Decentralised Centralised

DataModel Key Value Sorted Map

API Get, Put Get, Put, Scan, Delete

Security No Access Control

Partitioning Consistent Hashing Key Range Based

Replication Successor Nodes in ring Chunk Servers in GFS

Storage Plugin SS Table in GFS

Membership and Gossip Based Protocol Handshakes

18
Conclusions
● Fast, Flexible non relational database
● Reliable performance and Automated scaling
● Great fit for mobile, web, gaming, ad tech, IoT, and many other applications.
● Cost Effective in some use cases.

19
DEMO

20
References
● https://www.uniassignment.com/essay-samples/information-technology/revi
ew-of-the-motivation-for-nosql-information-technology-essay.php
● https://cloudacademy.com/blog/amazon-dynamodb-ten-things/
● https://aws.amazon.com/dynamodb/
● https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Get
tingStarted.Java.html
● https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Jav
aDocumentAPICRUDExample.html

21
Questions???

22