FIRST CHAPTER

Página 1

FACTS - CLASS 2

- There are 2.5 quintillion data per day being generated;

- 90% of the data generated on the planet was generated in the last two years;

- 80% of data is unstructured or in different formats;

- The 4 V's of Big Data -> Volume, Variety, Speed, Veracity;

- Volume: Data Size;

- Variety: Data Format;

- Speed: Data Generation;

- Veracity: Data Reliability;

- Big data and Data Science are not the same things.

- Big Data is the raw material x Data Science is a set of techniques for analyzing data.

- When applying Data Science to Big Data, Big Data Analytics is obtained.

- Data Engineer -> Extract and store all data

- Data Scientist - Process and analyze

Página 2

Storage System - Class 3

- It is important in storing data to know how it will be accessed.

- structured data is used data warehouse.

- unstructured data is used data lake or data store.

- Relational Database has schema defined before data storage

- Data Warehouse is created by some technology DBMS - Database Management System or (SGBD in Portuguese).

- NoSQL Database cannot be unstructured data.

- Data Warehouse - stores a large amount of data from different sources.

- The goal is to feed business intelligence (The goal: Alimentar).

- DW Benefits: Better Business Analytics, Faster Queries, Data Quality Improvements, Historical View.

- Data Lake Benefits: Raw Format Storage, Real-time Data Import.

- Data Store Benefits: Flexibility.

Página 3

Parallel Storage and Processing - Class 4

- Cluster a set of computers that form a server

- Parallel Storage - Distribute storage across multiple servers

- Hadoop HDFS- Hadoop Distributed File System cluster manager

- You can build a Data Lake that runs on a Cluster and allows you to store large volumes at a low cost.

- Job Tracker manages the processing

- Task Tracker does the work in the process

Página 4

Cloud Computing - Class 5

- Delivery of services over the internet

Página 5

Machine Learning - Class 6

- Area that focuses on using data and algorithms to mimic human learning.