Case Study on AWS SQS
HOW BMW USING AWS SQS AND MADE INVENTION OF CAR-AS-A-SENSOR
Let us understand in very detail what AWS SQS is.
Visualization is very important if we really want to understand in-depth. So, in this article, rather than just text, it will include many pictures and then explanations to understand the core meaning.
This article will be divided into several phases which will have its link with the former one:
> Need of Program called MQ
If we comprehend the why phenomena, the question of “what” is automatically answered.
So, why we need SQS, lets discuss scenario — use case — consequence — solution.
I have an Amazon shopping account — and the requirement is to but an item:
What are the steps involved?
- Choosing a product
- Adding in the cart
- Pay the bill
- An email comes for the confirmation
There may be some other steps involved but generally, these events occur.
Use case: So, it's like there is one program named“P1” managing the cart info, then another program “P2” to process the payment, and third program “P3” to send a confirmation email to the customer.
Consequence: what if the last step for sending the mail repeats several times, the customer wouldn’t like this experience and the company’s feedback is demolished.
What if the customer adding just only once the product in the cart, but 10 same products get added, again reputation is on the line.
Currently, we can say that P1, P2, P3 are tightly coupled.
So, maybe we need one more program say “P4” to monitor all the changes done by the customer and this program will receive the message and other programs (like for sending emails and so on) need to retrieve the info from P4 and then perform the tasks accordingly.
Solution: The program P4 is the Messaging Program also known as MQ.
-> What is SQS?
n The MQ program stores message for consumers to retrieve(technically known as Poll). One of the MQ programs is AWS SQS.
USE CASE OF SQS IN INDUSTRY.
BMW
BMW is using AWS to build a cloud service car, where sensors will use the tonnes of data hosted on the Cloud.
In AWS re-invent 2015:
Kai Demtröder — Vice President of Data Transformation, Artificial Intelligence, Data and DevOps Platforms
BMW Group
He says in the launch of BMW series 7 “because of services like S3, Elastic Beanstalk, SQS we were able to cope up with the requirements”.
And the cloud watch services the chart is prepared for performance efficiency. portal view and architecture view.
WHY BMW being a global infrastructure of premium automobiles, has to associate with AWS?
To generate digital innovations in 2015, BMW created a centralized database, to collects and combines data from sensors in vehicles, data warehouses, operational systems, and predictive insights.
However, this innovation was affected and slowed down by its own IT infrastructure and then decided to collaborate with cloud service providers. AWS stands best among them with respect to the market share it has compared to all cloud services.
What inventions does BMW make with the help of AWS service?
BMW built its new car-as-a-sensor (CARASSO) service in only six months leveraging Amazon Simple Storage Service (Amazon S3), Amazon Simple Queue Service (Amazon SQS), Amazon DynamoDB, Amazon Relational Database Service (Amazon RDS), and AWS Elastic Beanstalk.
What is the architecture used in CARASSO?
It adapts to rapidly changing load requirements that can scale up and down by two orders of magnitude within 24 hours. By 2018 CARASSO is expected to process data collected by a fleet of 100,000 vehicles traveling more than eight billion kilometers.
A car has been connected to world map data through cloud service.
Service used in this architecture.
- Route 53 — to connect to the core DNS database which helps in accessing any host and DNS name.
- Then infrastructure as s service VPC is used to configure all the services and subnets (JOB SERVER, MANAGEMENT SERVICE, RECEIVER SERVLET, WORKER SERVLET) inside it.
- Further, S3 is used in this VPC to store the huge bulk of data — Bigdata
- Elastic Load balancer, DynamoDB, RDS, Worker Queue SQS, S3
- RDS and DyamoDB databases are used for management purposes to monitor and all the real-time data.
- Queue SQS is used here for decoupling RECIEVER and WORKER SERVLET. The receiver keeps on adding messages to the SQS and Worker keeps on polling ( downloading the message, process it, and then copy it to the RDS database and MapData.
- Since the RDS database is linked to the Management servlet, they monitor the current data trends, and accordingly, the production decision is taken.
- Also, the message is copied in GLOBAL MAP DATA, and this is a place where our car is linked and CARASSO comes into play.
The working is SQS here is same as we discussed above case scenario with the program P1, P2, P3, P4.
I hope, this case study would have helped you understand how MNC’s like BMW is using AWS services and inventing their own services “car-as-a-sensor”
Thanking You…..
