NOTE This file was generated by Descript 1 00:00:00,050 --> 00:00:01,580 Hi guys and welcome back. 2 00:00:01,609 --> 00:00:04,987 In this video, let's talk about SQL databases and Mongo 3 00:00:04,987 --> 00:00:06,577 DB and how they are different. 4 00:00:07,054 --> 00:00:10,264 The usual databases are built from tables. 5 00:00:10,874 --> 00:00:15,584 SQL databases that use the SQL query language which we will learn a little 6 00:00:15,604 --> 00:00:18,379 bit about in a moment use tables as well. 7 00:00:18,379 --> 00:00:20,239 So they are the usual databases. 8 00:00:20,641 --> 00:00:24,841 In SQL databases, you build relationships between tables via 9 00:00:24,841 --> 00:00:27,811 the use of primary and foreign keys. 10 00:00:28,321 --> 00:00:33,361 In this table, which describes users, the id column would be the primary key. 11 00:00:33,811 --> 00:00:36,901 And you could have something like their bank accounts which has a 12 00:00:36,901 --> 00:00:38,901 foreign key which is the user_id. 13 00:00:39,331 --> 00:00:44,521 So with SQL, with the query language, you could then get information from 14 00:00:44,521 --> 00:00:47,131 both tables by using this relationship. 15 00:00:47,634 --> 00:00:51,954 Even though the relationship may not necessarily be, you know, hard coded 16 00:00:52,134 --> 00:00:56,934 in the code that relationship exists because the user ID in one table should 17 00:00:56,934 --> 00:00:59,184 be the same as the ID in another table. 18 00:00:59,274 --> 00:01:03,564 So you can use SQL, which looks, something like that to select information 19 00:01:03,564 --> 00:01:04,824 from both tables at the same time. 20 00:01:05,341 --> 00:01:07,711 But NoSQL is a bit different. 21 00:01:08,251 --> 00:01:13,031 In NoSQL, we don't use the SQL query language, which is the code 22 00:01:13,081 --> 00:01:14,431 that you just saw a moment ago. 23 00:01:14,994 --> 00:01:19,104 Instead the different databases, mongo DB is one of them, use 24 00:01:19,104 --> 00:01:21,354 different ways of querying that data. 25 00:01:21,868 --> 00:01:25,108 Also databases store data in different ways. 26 00:01:25,468 --> 00:01:30,808 And this is because each way of storing data provides some different benefits. 27 00:01:31,393 --> 00:01:34,783 However each way of storing data also has some drawbacks. 28 00:01:34,993 --> 00:01:40,693 So you can pick a different NoSQL database depending on what benefits you want and 29 00:01:40,693 --> 00:01:42,793 what drawbacks you're willing to accept. 30 00:01:43,332 --> 00:01:46,022 MongoDB doesn't use tables with columns. 31 00:01:46,172 --> 00:01:50,582 That are some NoSQL databases that use tables with columns but not very many. 32 00:01:51,122 --> 00:01:54,572 Instead MongoDB uses collections with documents. 33 00:01:54,902 --> 00:01:58,862 And you might be thinking, okay, that's just semantics surely they are the same. 34 00:01:59,222 --> 00:02:00,302 But they're not quite the same. 35 00:02:00,752 --> 00:02:03,962 The key difference is that when you have a table with columns all 36 00:02:03,962 --> 00:02:06,632 the rows contain the same fields. 37 00:02:06,932 --> 00:02:10,382 Let's say you have three columns and you have three rows. 38 00:02:10,562 --> 00:02:14,012 You know that the three rows each have three values in them. 39 00:02:14,552 --> 00:02:17,749 However, MongoDB documents can contain whatever you want. 40 00:02:18,439 --> 00:02:22,459 So they don't have to have the same stuff in them as other 41 00:02:22,459 --> 00:02:24,349 documents in the same collection. 42 00:02:24,499 --> 00:02:27,642 So your collections could be made up of disparate things. 43 00:02:28,199 --> 00:02:29,759 Now that might look a bit weird. 44 00:02:29,759 --> 00:02:33,769 It might feel wrong but it's one of the key elements of MongoDB. 45 00:02:33,869 --> 00:02:35,039 It's that flexibility. 46 00:02:35,429 --> 00:02:37,439 So what does a document look like? 47 00:02:37,469 --> 00:02:38,639 If it's not a table. 48 00:02:39,142 --> 00:02:41,692 You document is actually something like this. 49 00:02:41,722 --> 00:02:43,702 It is basically JSON. 50 00:02:44,092 --> 00:02:47,302 And if you don't know what JSON is, then you're looking at it right now. 51 00:02:47,382 --> 00:02:50,992 JSON is a collection of field names and values. 52 00:02:51,502 --> 00:02:55,822 The ID or underscore ID field is a required field inMongoDB documents. 53 00:02:56,257 --> 00:02:58,927 You can have things like embedded sub documents. 54 00:02:59,317 --> 00:03:01,117 And this here is a field. 55 00:03:01,147 --> 00:03:05,260 You can see the name goes before the colon and after the colon, you've got the value. 56 00:03:05,757 --> 00:03:10,257 The embedded sub documents are just documents that are a value 57 00:03:10,257 --> 00:03:12,027 to a field in another document. 58 00:03:12,527 --> 00:03:17,777 You can see that the values of the fields can be strings or integers but there's 59 00:03:17,777 --> 00:03:20,207 also a bunch of other types of value. 60 00:03:20,297 --> 00:03:25,217 For example, the underscore ID field has a value of type object ID which 61 00:03:25,217 --> 00:03:26,807 is something that's unique to MongoDB. 62 00:03:27,167 --> 00:03:30,767 And that identifies the different documents uniquely. 63 00:03:31,264 --> 00:03:33,814 We're going to learn more about what sort of stuff we can store in 64 00:03:33,814 --> 00:03:35,674 MongoDB as we go through the course. 65 00:03:36,151 --> 00:03:40,561 So I said earlier that NoSQL databases have benefits and drawbacks. 66 00:03:40,561 --> 00:03:42,731 Let's talk about some of the benefits of MongoDB. 67 00:03:43,237 --> 00:03:46,027 The first one is you don't have a fixed schema. 68 00:03:46,057 --> 00:03:48,727 You don't have these fixed columns in each table. 69 00:03:48,847 --> 00:03:50,437 So you have a bit more flexibility. 70 00:03:50,437 --> 00:03:53,737 You can just store whatever you want inside each document. 71 00:03:54,264 --> 00:03:58,254 And this can lead to faster development times because you don't have to worry 72 00:03:58,254 --> 00:04:02,761 about changing your database every time you want to add a different column or 73 00:04:02,761 --> 00:04:05,071 a different field to your documents. 74 00:04:05,251 --> 00:04:08,521 So you can just go ahead and do it without having to go into your database, 75 00:04:08,521 --> 00:04:13,491 make changes there and then restart your database or potentially delete other rows 76 00:04:13,491 --> 00:04:15,441 that didn't have those fields, et cetera. 77 00:04:15,996 --> 00:04:19,326 The data storage in MongoDB, which is JSON, is very similar 78 00:04:19,326 --> 00:04:20,586 to Python dictionaries. 79 00:04:20,766 --> 00:04:25,146 So when you extract data from MongoDB, it's very easy to automatically 80 00:04:25,146 --> 00:04:26,676 convert it to Python dictionaries. 81 00:04:26,856 --> 00:04:32,706 And you don't need to map data from the database into a native data structure. 82 00:04:33,231 --> 00:04:36,651 Also something I particularly like about Mongo DB is the scalability 83 00:04:36,651 --> 00:04:38,301 model is quite straightforward. 84 00:04:38,481 --> 00:04:42,981 So when you reach the limit for what a single MongoDB database can 85 00:04:42,981 --> 00:04:47,781 do, it's relatively easy to use two databases that are linked together. 86 00:04:48,317 --> 00:04:51,677 Although we won't get to this part of MongoDB in this course, it's something 87 00:04:51,677 --> 00:04:53,177 that you can do relatively easily. 88 00:04:53,397 --> 00:04:58,187 MongoDB University has some free courses that you can take on MongoDB scalability 89 00:04:58,307 --> 00:04:59,237 and they're pretty good as well. 90 00:04:59,767 --> 00:05:05,194 That are drawbacks to MongoDB and the main drawbacks are the main strengths of SQL. 91 00:05:05,734 --> 00:05:08,494 SQL is popular and it is popular for a reason. 92 00:05:08,991 --> 00:05:13,801 The first one is that SQL has JOINs and MongoDB does not have JOINs. 93 00:05:14,316 --> 00:05:17,136 What that means is using SQL and a SQL database. 94 00:05:17,256 --> 00:05:20,406 You can select data from two or more tables at the same 95 00:05:20,406 --> 00:05:22,776 time using the database code. 96 00:05:22,986 --> 00:05:24,606 And that is very fast. 97 00:05:24,906 --> 00:05:26,736 In MongoDB, you can't do that. 98 00:05:26,736 --> 00:05:31,326 You have to select manually using the Python code and therefore the 99 00:05:31,326 --> 00:05:35,386 selection of two tables or more at the same time is much slower. 100 00:05:35,882 --> 00:05:40,022 MongoDB can use more memory because you have to store the key name 101 00:05:40,022 --> 00:05:41,852 for each field in every document. 102 00:05:42,032 --> 00:05:46,202 Whereas when you use SQL, the column names are only stored once for each table. 103 00:05:46,707 --> 00:05:49,567 As usual there are limits to the size of documents and the amount 104 00:05:49,567 --> 00:05:50,737 of nesting within documents. 105 00:05:50,947 --> 00:05:53,377 But that is also true of SQL databases. 106 00:05:53,377 --> 00:05:55,927 So this is not a big drawback, but it's something to take into account. 107 00:05:56,445 --> 00:06:00,465 Something that you can run across when using MongoDB is that if you store 108 00:06:00,735 --> 00:06:05,942 a lot of high size things in your documents, MongoDB slows down massively. 109 00:06:05,972 --> 00:06:10,442 For example, if you put PDFs or images in your documents that can result 110 00:06:10,442 --> 00:06:13,118 in much slower speed for MongoDB. 111 00:06:13,627 --> 00:06:17,277 This is not a comprehensive list of the benefits and drawbacks of MongoDB. 112 00:06:17,407 --> 00:06:18,517 These are just the main ones. 113 00:06:18,757 --> 00:06:23,737 But there are other things that MongoDB has as edge cases that you should be 114 00:06:23,737 --> 00:06:28,297 aware of when you start using MongoDB in a high performance environment. 115 00:06:28,825 --> 00:06:31,375 However, I can tell you that I've used MongoDB in a high-performance 116 00:06:31,375 --> 00:06:33,655 environment and it works very well. 117 00:06:33,745 --> 00:06:37,405 So it's always going to come down to exactly what you're doing with your code. 118 00:06:37,405 --> 00:06:40,565 Whether you should choose MongoDB, or a slightly different NoSQL 119 00:06:40,585 --> 00:06:43,285 database, or even a SQL database. 120 00:06:43,762 --> 00:06:44,152 All right. 121 00:06:44,152 --> 00:06:45,832 Thank you guys for joining me in this video. 122 00:06:45,982 --> 00:06:48,412 I hope you've learned something and I'll see you in the next one.