- By Prachi Bhatnagar
- 80% of the online population has used the internet to make a purchase, and 50% of the online population has purchased online more than once.
- The E-Commerce sales are growing by more than 19% a year and will be of almost $1.4 trillion by 2015.
- 71% of shoppers believe they’ll get a better deal online than in stores
- E-retail will grow to 9% in 2016, up from 7% today.
These Facts seem pretty interesting to discover more about E-Commerce Industries and might intrude you to read about their technical difficulties and components.
So further conferring to a report by Forrester.com, nowadays, global ecommerce footprint has turned out to be the end goal. To reach that goal, companies are taking a strategic look at the opportunities around the globe and balancing the size of each market with factors such as ease of entry and the competitive landscape.
As a matter of fact, eCommerce corporations twinge with the online traffic during crowning times, allied concerns such as Scalability, Flexibility, Single-Point Failure, High Availability and added downsides of Traditional Computing.
One such company titled as Amazon, the world’s biggest E- Commerce Company known for its technical excellence came across such difficulties leading to develop something that would decentralize the infrastructure by providing services to development teams which supplemented to Amazon Web Services. Thus, AWS became a supportive explanation because of its performance advantage and cost competitiveness, greater durability and availability, less operational friction and strong data security.
Companies like CASA & VIDEO, ZAGG INC., ATG MEDIA, CHEFTV, etc. used the AWS Cloud components to outplay these problems. There are many ways in which eCommerce companies can use AWS Infrastructure components, including their web front end, marketing infrastructure, recommendation engines, etc. Some of the sample architecture as available on AWS Architecture Center are,
1. Web Front End – This is the first part of any e-Commerce site that is exposed to a customer & customer expects quick load times, seamlessly browsing through multiple product categories, with relevant information like images, data sheets etc. They also expect to seamlessly search for products using different keywords. Managing such infrastructure which delivers high performance can be a worrisome task & is one of the important factor to keep customers happy. This sample architecture offers one of the many ways in which AWS components can be used to build a highly scalable & fault tolerant web front end for eCommerce sites.
- A customers journey starts by typing the url and these requests can be handled by Amazon Route 53, a highly available DNS service.
- Next is the page load and this has to be highly optimized for load time, Amazon CloudFront can be used to ensure quick delivery of images from the nearest edge location & reduce latency. The source of these images & other content is S3 (Simple Storage Service).
- During peak times, infrastructure should seamlessly scale. Use of Amazon Elastic Beanstalk is recommended as it automatically handles capacity planning, load balancing, auto scaling & health monitoring.
- Other AWS Components like DynamoDB can be used for to store persistent data like the shopping cart, ElastiCache can be used in conjunction with DynamoDB as a caching layer to reduce I/O & improve performance.
- All the searches on e-Commerce website can be powered by Amazon CloudSearch which offers search in 34 different languages, allows Geospatial search & supports various data types. To Download the Diagram Click here.
2. Checkout Pipeline – After a customer has browsed through & made few selections, he then moves to the checkout process to confirm those purchases.
- The Check-out application has a separate Elastic BeanStalk which deploys all the applications and order of the workflow once the user reaches the Payment Gateway.
- It entails Amazon SWF which coordinates with all running order workflows by using SWF Deciders and SWF Workers for various purposes.
- The SWF Deciders and SWF Workers are triggered, once the customer instigates on filling the details. These are triggered for serving the purpose of verifying the particulars.
- Check-Out Service also involves Amazon SES for sending out transnational emails like, confirmation mails about the purchases that a user has made and Amazon RDS to store customer and orders databases. To check out the architecture in detail click here.
3. Marketing and Recommendations – Recommendation engine is one of the most important bits in the eCommerce architecture as it helps the customer find products which are similar to his / her liking. Marketing emails can be triggered based on the search history of the customer.
- This contains user profiles which fetches the database based on previous searches from Log File Repository through Amazon EMR.
- It has two separate Elastic BeanStalks – one to manage Marketing Management App which is used to send mails to the customers and the other to manage Recommendation Web Service which shows the user information based on the previous searches.
- Marketing And Recommendations also involve Amazon RDS to fetch customer database for E-Mail Campaigns and DynamoDB for Recommendation search. Check out the full architectural Diagram here.
AWS helped the firms reduce the time necessary to introduce improvements to the market, while providing a level of redundancy that is difficult to achieve with traditional infrastructures.
Have some suggestions on the blog? Share it in the comments or give us a buzz.