Okay, let's quickly cover the nine core features of BigQuery and then explain a little bit of the difference between data scientists, data analysts and data engineers. And we'll get you launched into your next lab. So BigQuery, the core tenet is, don't manage the infrastructure yourself. So if we haven't made this clear enough, if you don't want to be buying hard drives and managing that hardware especially. Even if you have it perfect today, a year from now when you're processing half that data or ten times that data, let the platform scale for you. And it allows you to focus on finding those insights yourself. Become really, really good, not as the jack of all trades, in managing your hardware, and writing the queries, and doing the job of ten different people. Become extremely deep and proficient at mining those insights. And if you wanted to double down and continue to take these additional courses and pick up how to do machine learning. It's, again, along this ramp of really focusing on those insights instead of infrastructure. Okay, so BigQuery in a nutshell. So nine key points here. So it's the fully managed petabyte-scale data warehouse. It's as you saw with the pictures, it's backed by Google data centers. The economies of the cloud mean that you pay for only what you consume, plus the cost of storage if you're creating those permanent tables. Next up, security. It is access controlled and you actually, within your Google Cloud project, can manage access for members and groups and whoever needs to actually have access to your datasets as well. And we'll cover a lot of those in one of the data access modules as part of the third course. So your data itself, if you're more concerned with how Google is storing it, is encrypted and transport in at rest for your data centers and replicated. Every query, as you saw within that dashboard in data studio, each of the different queries is actually logged as a separate transaction. If you're familiar with Stackdriver, you can actually monitor all those queries that are going through and then look back for that history as well. And the big benefit, as we mentioned, is scalable, Bigquery. So you can process multiple queries in parallel, actually up to 50 concurrent queries at the exact same time, going on at once. Last points, with a little bit of SQL you can use joins and unions and bring your data together across many different data sets and really break apart those data silos. Number eight, the reason why you're hopefully taking this course is you want to get better with your SQL, because BigQuery loves when you write excellent, awesome SQL to get those insights out of it. And another interesting point, if you really like the behind the scenes architecture, is BigQuery actually has no indexes and even actually has no keys, if you're familiar with database terminology. So again, as part of an analytics warehouse, that architecture and performance discussion is something that's very, very fun to dive into. And again, it's both on open standards as well. Explore around and find pre-built queries and examples online of over 50 different data sets that are available for you to explore as well. So as I like to say, good artists copy, great artists steal. So if you find a query that looks great that somebody else has written, steal it, modify it, you know, make it work for you. All right. So we mentioned the three different ways to access BigQuery. Primarily going to be focused on the web UI for this specialization. You can also access it through the command line, which is great. And then much like if you wrote any other application, you can get your queries ran over the web through RESTful APIs.
