The widening of the customer base is the biggest challenge facing leading companies, but Amazon’s integration of big data analytics into its business strategy has helped to set trends and better adapt to contemporary developments. It also left a more positive impact on the level of financial revenue and reduced costs but extended to create a strong competitive advantage and a broader vision about its future aspirations.


hi 2016, Amazon recorded more than 304 million active client accounts, 65% of which were female, and received an average of 4,310 visits per minute. Amazon Prime members spend an estimated $1,500 on average, while other customers spend about $625 annually; in addition, members have access to over 1 million e-books.

Statistics show that the company has reduced prices for year-end offers (2015) for 489.5 million products, and acknowledged that 70% of purchases made by smartphones. Amazon’s market share during Black Friday’s was around 35.7%, while shipping costs were estimated at $1.54 billion, which makes the company actually seeking to gain the best by looking for mechanisms to integrate and exploit the entrance of big data analytics to push and develop its strategy and make a qualitative leap in the field of electronic commerce [18].


AWS provides a wide range of managed services to quickly and easily create comprehensive, secure data applications, whether applications require physical flow or aggregated data processing.

AWS provides the infrastructure and tools to process a big data project by collecting, storing, and analyzing it. AWS has an ecosystem of analytical solutions specifically designed to handle this growing amount of data and provide a comprehensive view of the company’s business model.

Amazon’s white paper lists the technical tools used to handle big data [19]:

  • • Amazon Kinesis: It is an AWS data-streaming platform, making it easy to download and analyze streaming data, and provides the ability to create custom data streaming applications for specialized needs. Using Kinesis, you can accommodate real-time data such as application logs, site clicks, IoT telemetry data, and more in databases, data lakes, and data warehouses, or create real-time applications using this data. Using streaming data allows real-time dashboards to be triggered, alerts generated, pricing, and advertising implemented Dynamic. Amazon Kinesis enables data processing and analysis as it arrives and responds in real-time, rather than having to wait until all data is collected before processing can begin.
  • • AWS Lambda: It allows code to run without providing or managing servers, the company pays only to calculate the time it consumes, and there is no charge when the code is not miming. With Lambda, you can run code for any type of application or backend sendee-all without management. The company engineer raises the company code while Lambda handles everything needed to run it and expand the code. The code can also be set up to ran automatically from other AWS sendees or connect to it directly from any web or mobile app.
  • • Amazon EMR: It is a highly distributed computing framework for processing and storing data quickly, efficiently, and at a low cost. Amazon EMR Apache Hadoop, an open-source framework, is used to distribute and manipulate data across a resizable set of Amazon EC2 instances and allows you to use the most common Hadoop tools such as Hive, Pig, Spark, etc. Hadoop provides a framework for running massive data processing and analytics. Amazon EMR does all the work involved in providing, managing, and maintaining Hadoop Group’s infrastructure and software.
  • • AWS Glue: It is a fully managed ETL service that can be used to reliably index, clean, emich, and transfer data between data stores. With AWS Glue, the cost, complexity, and time it takes to create ETL functions can be greatly reduced.

In another white paper for 2019, Amazon added [20]:

  • • Amazon Machine Learning: It is a sendee that makes it easy for anyone to use predictive analytics and machine learning technology. Amazon ML provides browsing tools to guide you by creating machine- learning models (ML) without learning complex ML algorithms and techniques. The embedded processors are guided through interactive data exploration steps and trained in the ML model to assess the model quality and adjust outputs to match business objectives. After the model is ready, predictions can be requested either in batches or by using a low real-time API.
  • Amazon Athena: It is an interactive query sendee that makes it easy to analyze data on Amazon S3 using standard SQL. There is no installation or management infrastructure. You can start analyzing data immediately, and you don’t need to upload your data to Athens, as it works directly with the stored data hi S3. It only requires logging into the Athena console, defining the table schema, and starting the query.
  • Amazon DynamoDB: It is a fast, folly managed NoSQL database sendee, which makes storing and retrieving any amount of data simple and cost-effective, and selves any level of application traffic. DynamoDB helps to unload the administrative burden of operation and expand the database in a more distributed and accessible maimer.
  • Amazon Redshift: It is a fast, large, and folly managed data warehouse storage sendee that allows you to analyze all data efficiently, effectively, and at low cost using BI tools. Optimized for data collection ranging from hundreds of gigabytes to beta bytes or more, designed to cost less than one-tenth of the cost of traditional data storage, this system automates most common administrative tasks associated with supplying, configuring, monitoring, backing up, and securing the data warehouse, making it easy to manage and inexpensive. This automation allows the construction of petabytes-sized data warehouses in minutes, rather than weeks or months taken by traditional local implementations.
  • Amazon Elastic Search: The flexible search makes it easy to deploy, operate, and extend searching for log analytics, foil-text scanning, application monitoring, and more. AES is a folly managed sendee that provides easy-to-use APIs for flexible search and enhances availability, expansion, and security according to the requirements of the production environment. The seivice offers implicit integrations with previous seivices to quickly move from raw data to actionable insights.
  • Amazon Quick Sight: It is a fast, easy-to-use business analytics sendee miming on the cloud, making it easy for all employees within an organization to create visualizations, conduct custom analysis, and quickly get business insights from their data anytime, on any device.

It can connect to a wide range of data sources, including flat files, for example, CSV and Excel to access logical databases, including SQL Server, MySQL, and PostgreSQL.

Organizations extend tlieir business analytics capabilities to hundreds of thousands of users and provide fast, responsive query performance using a powerful in-memory engine (SPICE).

Ian et al. also added [21]:

• Amazon EC2: It provides an ideal platform to run large self-managed data analysis applications on AWS infrastructure. Almost any software can run on Linux or Windows virtual environments on Amazon EC2, and you can use the Pay As You Go pricing model, and there are many options for self-managed big data Analytics.

< Prev   CONTENTS   Source   Next >