Downtimes & Failures every Holiday Season
Downtime hassles and other technical difficulties like Single point of failure, Power Failure and further security issues is a burning topic for E-Commerce websites during the Holiday Season. Specially on Thanksgiving, Cyber Monday, Black Friday and Christmas when eCommerce industry clocks a revenue of over 25% of their annual revenues.
The complication of partial and complete downtime for various E-commerce websites is a persisting problem owing to increased online traffic during crowning times. This further leads to ruthless imprint on customers and an increased competition from the corporations that are performing better.
A recent study based on the facts of internet retailer magazine at the end of 2013 shows the e- commerce websites that suffered downtime hassles which continued to be a disappointment for the customers shopping online.
Some who perished & disappointed their Customers
Some of these websites that were a newscast during the holiday season in 2013 were:
1. Motorola – As part of Cyber Monday, Motorola announced that it would offer an off contract and customizable version of the Moto X for $150 off its regular price. But once the online shopping day rolled around, Motorola was hit with a surge in traffic that caused the website to crash.
2. Wal-Mart – The site went down for a brief period, about an hour and was very slow through most of Black Friday leaving customers frustrated.
3. Belk – Customers rushed to the company’s site with the intent to capitalize on several Black Friday deals, many were frustrated with the repeated crashes.
4. Future Shop – One of Canada’s largest consumer electronics retailer, was also ill prepared for the holiday shopping traffic. As reported on Black Friday, traffic overload likely tied up the server, making the site unreachable for visitors.
5. RBS – Thousands of Cyber Monday shoppers struggled to complete their check-out process after an IT glitch at RBS hit banking transactions.
On the contrary there were other E-commerce websites that were ranked amongst the top by the Internet Retailer Magazine which included websites like Amazon.com, Ice.com, etc.
To further deliberate on these websites, it is essential to rug light on AWS services of High- Availability Architectures, Design for Failure, and AWS Building Blocks which help in eliminating the risk of single machine outages, data center outages, network hiccups, and more.
Design for Failure & High Scalability
Collective key solutions provided by Amazon Web Services (AWS) to moderate these issues of downtime errors for E-Commerce websites contain:
- Inherently Highly Available & Fault Tolerant: AWS Services like S3, CloudFront, SNS, Route 53 and ELB are inherently fault tolerant, robust and scalable in nature.
- Highly Scalable, Available & Fault Tolerant with Right Architecture: AWS Services like EC2 when used appropriately with Elastic Load Balancer & AutoScaling can help in automatically increasing the server capacity, detecting & replacing faulty EC2 services.
Right usage of these building blocks can eliminate the chances of downtime and help achieve 99.99% available while your infrastructure scales up & down as per the traffic on your web store.
Multi AZ Architecture
Additionally, Amazon Availability Zones are distinct physical locations having Low latency network connectivity between them inside the same region and are engineered to be insulated from failures from other AZ’s. They have Independent power, cooling, network and security.
Using AWS, companies can build infrastructures as secure as, and possibly more secure than, those they can build on-premises.Use of multi-AZ architecture drastically reduces the chance of failure owing to problems at one particular location within the AWS infrastructure. User traffic can automatically be directed to another zone in case of failures & avoid any disappointed customers.
Loosely Coupled Systems:
Loosely coupled systems are more fault tolerant and can achieve a higher scalability. Using AWS services like Simple Queue Service helps in designing loosely coupled systems.
1) Using queues (Amazon SQS) buffers against failures & isolate components
2) Design every component such that it expose a service interface and is responsible for its own scalability and interacts with other components asynchronously
3) Bundle the logical construct of a component into an Amazon Machine Image so that it can be deployed more often
4) Make your applications as stateless as possible.
Some who Weathered the Load
Substantiating to the solutions are numerous case studies by AWS, companies that were saved from downtime and other difficulties. Some websites included were Zagg Inc., Futbol Club Barcelona, Ice.com and many others.
1. ZAGG, Inc. produces protective coverings for smartphones, tables, notebooks, and other small screen devices. The company established its site using the LAMP stack on an Amazon Linux AMI in Amazon Elastic Compute Cloud (Amazon EC2). The infrastructure also includes a wide assortment of additional AWS features, including Elastic Load Balancing (ELB) and Auto Scaling for Amazon EC2, Amazon Elastic Block Store (Amazon EBS), Amazon Simple Storage Service (Amazon S3), and Amazon Relational Database Service (Amazon RDS), Amazon Route 53, Amazon CloudFront and Amazon Simple Email Service (Amazon SES)
2. Futbol Club Barcelona. The development team manages short message service (SMS), e-commerce, social media, and mobile apps. Amazon Web Services usage includes- Amazon S3, Amazon EC2, Amazon SNS, Amazon Route 53, Amazon CloudFront, Amazon CloudWatch, Amazon RDS and Amazon CloudFormation.
3. Ice.com : Since its launch, Ice.com has been voted best of the web by Internet Retailer magazine. Based in Champlain, New York, Ice.com was one of the first companies to offer jewelry online. Its usage of Amazon Web Services like Amazon EC2, Amazon EBS, Amazon ELB, Amazon CloudWatch, Amazon VPC, Amazon S3, AWS Identity and Access Management helped the website to be voted into the Hot 100 e-commerce companies by Internet Retailer in both 2012 and 2013.
Some Related Reads: