# NodeJS streams to solve a data movement problem, minimizing memory and increasing speed Date review: Mar 8, 2022 --- A NodeJS Stream system to solve a data movement problem | Case Study - YouTube As you may know, YepCode uses **JavaScript** as programming language, and **NodeJS** is the current execution engine. We are working to support other languages like **Python** or **Go!** in the future. NodeJS is an open-source, cross-platform, single-threaded **runtime environment** for building **fast and scalable** server-side and network **applications**. It uses a non-blocking, event-driven I/O architecture, making it efficient and **suitable for real-time applications**. One of the advantages of using NodeJS is that it allows you to **handle a large number of simultaneous connections with high performance**. It is fast, efficient and widely used for **Chatbots**, **Streaming** applications (Netflix uses it!) and **IOT**. As seen before, moving information between applications can be an extremely tedious task, especially if [**we are dealing with large volumes of data**](/content/using-yepcode/how-to-automate-efficiently-data-movements-from-google-bigquery-to-snowflake/index.html). Today we bring you an example demonstrating how YepCode can solve a data movement problem while **minimizing memory consumption** and **with great transfer data rates.** ## What can I use a NodeJS Stream for? A Stream is a flow of information that programmers usually use to **transfer data**. ‘Streams’ were built to **handle data in real-time** through a buffer. Their main advantage is that **you do not need to store the entire data in memory** at the same time, but parts of this data. So it can also process file streams that can be read in parts without their full context. Streams are very useful when we need to **query information** that comes **from multiple data sources**. This sample shows one need that one of our clients had, as they needed to [**download a CSV file**](/content/docs/libraries/request/index.html), [**process each CSV line**](/content/docs/libraries/csv/index.html), convert each entry to a JSON format upload it to an [**FTP server**](/content/docs/integrations/sftp/index.html) and also insert each row in a [**Microsoft SQL Server**](/content/docs/integrations/mssql/index.html) database. This use case could be a full project created from scratch, but with YepCode, and some dozens of lines of code, it’s done! The YepCode **integrations used have streams support**, so we are able to create a stream from the file URL, **pipe** the content to a transformer that **converts each entry to a JSON**, and in parallel, **pipe that JSON to the FTP and SQL Server integrations**. All the process is extremely **fast** and with no memory consumption! ## Watch how to execute this Stream process in the next video If you watch the video below, you will see an example of the execution of this process. As you may see, we copy the **URL** and when we hit the execution, **it already asks us for the URL of the file** that we include. We run it and this is already processing all the data. As we have said before, the process does not load all data in memory. You can also observe that the **process has finished successfully**, **processing** the entire **89 rows**. If you now go to the **FTP server** we will see that **the generated JSON data is here**. Here you can also see that the process has converted the **.CSV to a JSON** format. And finally, if you go to the **SQL Server Database**, you will see that we have 89 records. Remember that our [**Docs platform**](/content/docs/index.html) includes every single detail to make the most out of YepCode, Enjoy the video and… Happy coding! 🧑‍💻