Data Lakes and Data Lake Analytics



What is a Data Lake?


‘Data is the new oil’. How many times have we heard this age old saying? Nowadays, every business is slowly becoming dependent on data. You need data about your business to monitor and improve performance, you need customer data to generate leads. How do we store all this data?


Data storage and analytics have always been a major challenge for corporations. Previously, Data used to be stored using Data Storage and Data warehouses. In Data storages, the data needs to be formatted with a schema and stored using rows and columns. Data warehouses are a larger repository of data where the data are already filtered for analysis purposes.


Data lake on the other hand is completely different from the two. In a Data Lake, data is stored without formatting in its natural or raw format. This could include files, images, video, audio, emails, documents, pdfs, semi-structured data, structured data in row and columns, anything. There is no need of processing or formatting the data before storing, eliminating a significant lead time. The Data is transformed before using it for analysis, visualization and machine learning purposes etc.



How has it made our lives easier?


Imagine this, you have a large retail business across multiple states with thousands of daily invoices. You are collecting customer data for business performance monitoring, retention and lead generation purposes but you are unsure about what sort of analysis or visualization you will require it for. In case you have a data storage or warehouse, you would have to arrange the data in columns and rows and process it for a particular output before storing.


However, since you are not quite sure of what analysis you are expecting, it is next to impossible to transform the data into any format hence you end up not having a proper data storage. With Data Lakes you are able to store the data as it is, enabling you to store the data even if you are unsure of the desired output. When you do have an output in mind, you can reach into your data pool and create data analysis and visuals as per your requirement.



What are the available Data Lake solutions?


Data lakes came into being to meet the needs of Google and other large enterprises for a cost effective and efficient data storing and processing solutions. Companies like Google, Microsoft, Oracle, Amazon, Teradata, Hortonworks, Zaloni, IBM are all now offering data lake solutions through their cloud platforms with very attractive subscription options.


What else should you know?


While Data Lakes are easy to deploy, it also needs cautious handling by the users and less frequent direct access. As large volume of data is stored in raw unstructured format, users have to be very skilled to make sense of the data. The end users must have clear understanding of the desired output and the required data they are looking for.



Data Lake Analytics


There are solutions such as Azure Data Analytics and AWS Purpose Build Data Analytics Services that allow you to work with big data such as Data Lakes. You can use this for your desired analysis and visualizations. These are on-demand, job-based analytics solutions that develop and transform data parallelly allowing you to process data on demand and scale instantly. Allowing you to focus on more strategic business decisions while saving you money compared to contemporary data storage solutions.

17 views0 comments