1
00:00:00,150 --> 00:00:01,470
‫Hey, guys, what's going on?

2
00:00:01,800 --> 00:00:08,640
‫So we have talked about acid, the four properties in a theoretical manner, but it's time to put them

3
00:00:08,640 --> 00:00:14,730
‫in practice and explain those four properties in an actual database setting.

4
00:00:14,880 --> 00:00:20,520
‫So I'm going to pick this, but you can do the same exercise with any database you want for some.

5
00:00:20,790 --> 00:00:27,360
‫I would luck to use Docker because this way I can try out many, many databases with just one command

6
00:00:27,360 --> 00:00:30,060
‫and I have spit up another database just like that.

7
00:00:30,060 --> 00:00:30,660
‫So let's go ahead.

8
00:00:30,660 --> 00:00:32,730
‫And it's been a database right here.

9
00:00:33,520 --> 00:00:37,260
‫I've got to keep this up because I'm going to need that to a terminals, but I'm going to open up a

10
00:00:37,260 --> 00:00:38,640
‫new one right here.

11
00:00:39,000 --> 00:00:45,360
‫And let's do a little bit about Puskás Container going to Docker Run.

12
00:00:45,870 --> 00:00:48,780
‫Let's call it doctorand.

13
00:00:50,110 --> 00:00:53,700
‫Let's give it a name, dash, dash name is going to call it acid.

14
00:00:54,340 --> 00:01:01,510
‫It's got acid and we get to do we don't really need to expose the port or anything because we're going

15
00:01:01,510 --> 00:01:05,890
‫to really go into the container itself, bash into the container.

16
00:01:06,520 --> 00:01:13,610
‫The next command is let's go ahead and detach that and then let's pull up Puskás 13.

17
00:01:14,020 --> 00:01:19,000
‫I think that's the latest thing before we run that actually we forgot something very important, which

18
00:01:19,000 --> 00:01:24,760
‫is that she was Griss password equal postscripts.

19
00:01:24,940 --> 00:01:30,550
‫You have to give it pulseless user password as go go ahead and run that just like that is going to take

20
00:01:30,550 --> 00:01:35,050
‫a little bit while in your end to load the image.

21
00:01:35,050 --> 00:01:37,520
‫But once you have it, you're ready to rock and roll.

22
00:01:37,540 --> 00:01:37,960
‫Let's go ahead.

23
00:01:37,960 --> 00:01:43,010
‫And so so let's make sure that the container is running just like that.

24
00:01:43,030 --> 00:01:43,710
‫Yes.

25
00:01:43,720 --> 00:01:45,490
‫Here is my PJI acid.

26
00:01:45,490 --> 00:01:46,450
‫It's running.

27
00:01:47,140 --> 00:01:52,950
‫They always make sure that it's running because you might have spilled something in the command or did

28
00:01:52,990 --> 00:01:55,020
‫something bizarre or something is wrong.

29
00:01:55,030 --> 00:01:55,250
‫Right.

30
00:01:56,080 --> 00:02:02,500
‫So go ahead and do it and get it into the acid.

31
00:02:02,830 --> 00:02:03,910
‫And what can I do?

32
00:02:03,920 --> 00:02:08,590
‫We will execute a physical because of the PostgreSQL command.

33
00:02:08,950 --> 00:02:17,580
‫And I want to log in as Deshu that Posterous user that just like that, we are in the process commensal

34
00:02:17,890 --> 00:02:18,100
‫now.

35
00:02:18,100 --> 00:02:20,020
‫We can execute beautiful stuff.

36
00:02:20,020 --> 00:02:23,320
‫We can create a database, we can play that create tables.

37
00:02:23,560 --> 00:02:29,110
‫So since I am in the Postgres database, I have a database called Polska, so I'm going to play up with

38
00:02:29,110 --> 00:02:30,140
‫that database right there.

39
00:02:30,150 --> 00:02:30,390
‫Right.

40
00:02:30,670 --> 00:02:39,490
‫I was going to do create a table called Products to Test at Tomasetti and I'll go ahead and create table

41
00:02:39,490 --> 00:02:40,780
‫products.

42
00:02:41,120 --> 00:02:43,390
‫Let's do a PID serial.

43
00:02:43,390 --> 00:02:46,870
‫So that's a little increment in my school, I think it's called.

44
00:02:47,470 --> 00:02:48,810
‫And sure.

45
00:02:48,910 --> 00:02:51,220
‫Let's make this the primary key.

46
00:02:52,440 --> 00:03:00,100
‫And I think we can just continue doing that stuff right and then the online name as a text boom.

47
00:03:00,540 --> 00:03:08,280
‫What else price is double our shorts, double our Flodden because I keep forgetting in different databases.

48
00:03:08,700 --> 00:03:11,020
‫When you work on many databases, you get confused.

49
00:03:11,430 --> 00:03:13,740
‫And the finally is inventory.

50
00:03:14,010 --> 00:03:17,310
‫How much inventory do we have in this product?

51
00:03:17,310 --> 00:03:19,440
‫And let's just use an integer here.

52
00:03:21,010 --> 00:03:22,420
‫And that should be it.

53
00:03:23,050 --> 00:03:24,530
‫Yeah, that's what I thought.

54
00:03:24,550 --> 00:03:25,650
‫It's not called double.

55
00:03:26,290 --> 00:03:29,080
‫It's got to be flawed, I think.

56
00:03:34,220 --> 00:03:38,000
‫I think it's double in my school or I keep forgetting.

57
00:03:39,870 --> 00:03:46,200
‫All right, now we have a product sale, let's go ahead and create a sales table so we can produce sales.

58
00:03:46,230 --> 00:03:47,280
‫Great table sales.

59
00:03:49,260 --> 00:03:52,260
‫Let's go ahead and do a sale I.D. That's a cereal.

60
00:03:52,260 --> 00:03:58,890
‫Also primary key and enter the primary key.

61
00:03:59,010 --> 00:04:02,880
‫That's the product ID I can create as a foreign key, but doesn't have to.

62
00:04:02,910 --> 00:04:05,210
‫So that's an integer, right?

63
00:04:05,400 --> 00:04:07,590
‫That's a product I.D. to draw.

64
00:04:08,860 --> 00:04:10,480
‫Boom, price float.

65
00:04:10,490 --> 00:04:19,150
‫We learned our lesson, so that's an entry in the sales, could be a product, could be sold in a different

66
00:04:19,150 --> 00:04:19,630
‫price.

67
00:04:19,630 --> 00:04:19,880
‫Right.

68
00:04:19,990 --> 00:04:23,700
‫But the current price of the product is in the product table.

69
00:04:24,070 --> 00:04:25,510
‫This is what was sold.

70
00:04:25,510 --> 00:04:32,280
‫That and then the quantity as also an integer, a boob.

71
00:04:33,130 --> 00:04:33,460
‫All right.

72
00:04:33,460 --> 00:04:37,950
‫Now we have two beautiful tables let's test out at Tomasetti.

73
00:04:38,500 --> 00:04:42,040
‫So the first thing, listeners populate some some rows on this table.

74
00:04:43,000 --> 00:04:47,320
‫So I was going to do insert into products.

75
00:04:48,790 --> 00:04:59,100
‫And let's give it a name, I don't know what are we insert name and a prize and inventory or simply

76
00:04:59,160 --> 00:05:01,000
‫say, I'm just going to do it one one.

77
00:05:01,000 --> 00:05:04,950
‫Royo let's call this a table at tables.

78
00:05:04,960 --> 00:05:09,310
‫Very confusing phone selling phones.

79
00:05:09,970 --> 00:05:14,940
‫And the price of the phone is nine nine nine nine nine nine nine.

80
00:05:15,580 --> 00:05:17,950
‫That's an expensive phone inventory.

81
00:05:17,970 --> 00:05:20,510
‫Let's say you have hundred units, then boom.

82
00:05:20,950 --> 00:05:21,580
‫And then.

83
00:05:22,950 --> 00:05:23,400
‫That's it.

84
00:05:23,670 --> 00:05:25,140
‫That's the only product we have.

85
00:05:25,520 --> 00:05:29,520
‫OK, so now let's test out our Atomos City.

86
00:05:29,760 --> 00:05:36,690
‫So I mean, if we did select store for our products, you can see that we have this row.

87
00:05:37,500 --> 00:05:39,700
‫But we're going to do is we're going to make a sale.

88
00:05:39,750 --> 00:05:42,300
‫We're going to sell, let's say, 10 phones.

89
00:05:43,310 --> 00:05:44,690
‫Like we sold 10 phones.

90
00:05:44,870 --> 00:05:46,040
‫What does that mean?

91
00:05:47,770 --> 00:05:56,250
‫The fear of walking with Demand-Driven designed that is one has scribed, but it's in the database,

92
00:05:56,260 --> 00:06:05,800
‫it's broken into multiple queries in a single transaction, and so a sale is often units.

93
00:06:05,800 --> 00:06:12,820
‫That means you have to deduct 10 units from the inventory and then you have to insert a row in the sales

94
00:06:12,820 --> 00:06:17,170
‫table saying that you have inserted that you cannot do that.

95
00:06:17,170 --> 00:06:21,250
‫And two transaction, you have to do it in an atomic transaction.

96
00:06:21,280 --> 00:06:23,290
‫That's why we explained in the video the lecture.

97
00:06:25,120 --> 00:06:26,740
‫So we've got to begin a transaction.

98
00:06:28,480 --> 00:06:34,200
‫And then we're going to the first transaction we're going to do is update products, right?

99
00:06:34,570 --> 00:06:35,680
‫What was the phone idea?

100
00:06:35,970 --> 00:06:40,000
‫Let's just do from products you might want to query or process.

101
00:06:40,000 --> 00:06:42,560
‫Idea number one, processed product line number one.

102
00:06:43,000 --> 00:06:45,910
‫So we're going to do is update products.

103
00:06:47,920 --> 00:06:49,840
‫Set name.

104
00:06:52,000 --> 00:06:59,890
‫A bread product, set, inventory, inventory, equal inventory, minus 10.

105
00:07:01,710 --> 00:07:03,500
‫Well, that's in this transaction.

106
00:07:03,840 --> 00:07:08,890
‫So now if I query that product again, you can see it's 90, right?

107
00:07:09,660 --> 00:07:10,930
‫Very, very critical here.

108
00:07:11,430 --> 00:07:13,620
‫So now if I go ahead and insert.

109
00:07:16,270 --> 00:07:17,230
‫Let's assume.

110
00:07:19,010 --> 00:07:20,590
‫I crashed right here.

111
00:07:22,030 --> 00:07:23,740
‫If I crashed right here.

112
00:07:25,140 --> 00:07:35,730
‫What will happen, you just lost 10 units from your phones that you did a really sold OK, and that's

113
00:07:35,730 --> 00:07:39,420
‫the idea of atomistic because you have to follow it up.

114
00:07:39,660 --> 00:07:39,880
‫Right.

115
00:07:40,620 --> 00:07:42,450
‫So, like, let's say I crashed right here.

116
00:07:42,690 --> 00:07:48,870
‫I don't know, just this killed the whole container just like that.

117
00:07:49,770 --> 00:07:56,160
‫If you do that without automaticity, it will be a disaster if you come back to the database.

118
00:07:56,700 --> 00:08:01,560
‫And then you said select star from product and you saw that it's a 90, but no.

119
00:08:03,000 --> 00:08:05,370
‫It says hundred y.

120
00:08:06,360 --> 00:08:13,050
‫Because of animosity, we started a transaction, we did an update, but we crashed, we exited fossickers

121
00:08:13,050 --> 00:08:15,280
‫in this case simulating a crash, right.

122
00:08:16,860 --> 00:08:24,990
‫That is the idea of at the most of all, the transaction have to be done in as an attempt to split those

123
00:08:25,200 --> 00:08:28,170
‫and those do the transaction again in the right way.

124
00:08:28,980 --> 00:08:31,470
‫Again, transaction doing an update.

125
00:08:32,580 --> 00:08:39,480
‫To them, victory, and then now I'm going to insert a sale and certain the sales process ideia.

126
00:08:42,000 --> 00:08:49,920
‫Price and quantity, and let's just do values.

127
00:08:51,630 --> 00:08:57,840
‫Number one, I sold it and I thought I sold it at nine nine nine nine nine Sam Price, they sold 10

128
00:08:57,840 --> 00:08:58,520
‫units room.

129
00:08:59,620 --> 00:09:04,960
‫And you can break it out into five minutes at this price and five minutes at this price will not.

130
00:09:06,360 --> 00:09:14,940
‫And as we did of that, now, clearly the sale of Sable's you can see that we have 10 beautiful units

131
00:09:14,940 --> 00:09:20,310
‫right here and my products are 90.

132
00:09:21,560 --> 00:09:24,320
‫So now if you some of those puppies up.

133
00:09:25,310 --> 00:09:28,700
‫You get a hundred, and that's called consistency's.

134
00:09:28,780 --> 00:09:33,530
‫We just explained atomos and consistency and the same concept.

135
00:09:34,430 --> 00:09:39,950
‫And this is only viewable in my transaction.

136
00:09:40,310 --> 00:09:47,600
‫So now if I went to a completely different terminal and tried to query again, I'm not I'm not committing

137
00:09:47,600 --> 00:09:47,810
‫yet.

138
00:09:47,810 --> 00:09:48,110
‫Right.

139
00:09:50,570 --> 00:09:52,560
‫And I do like to do the same thing.

140
00:09:52,580 --> 00:10:01,640
‫Hopefully we got a school there, a big acid, then if I do select star from products, you can say

141
00:10:01,640 --> 00:10:05,570
‫they're still 100 and if you call it the sales, stable.

142
00:10:07,670 --> 00:10:16,490
‫There is nothing and that's part of isolation, because now we're we're in an atomic manner, so these

143
00:10:16,700 --> 00:10:24,430
‫two things were inserted in that transaction, but we didn't roll back or commit the moment I commit.

144
00:10:25,340 --> 00:10:36,170
‫Now, we have a nice view and out of the data and the data is now flushed and available, visible to

145
00:10:36,170 --> 00:10:36,800
‫all transaction.

146
00:10:36,810 --> 00:10:42,090
‫Now, if I created it again, you can see that we have a sales now zero.

147
00:10:42,090 --> 00:10:42,830
‫So that is one.

148
00:10:43,070 --> 00:10:45,620
‫And if you query the product table, you can see that's 90.

149
00:10:46,370 --> 00:10:54,230
‫So what I was to talk about here, talking about Midcity City, we talked about consistency because

150
00:10:54,740 --> 00:10:57,410
‫if you don't have to Midcity, you don't have consistency.

151
00:10:57,830 --> 00:11:04,220
‫And you can argue with that, obviously, guys, because you can say, hey, I can I can be smart and

152
00:11:04,220 --> 00:11:06,020
‫make a single query.

153
00:11:06,020 --> 00:11:06,270
‫Right.

154
00:11:06,290 --> 00:11:11,360
‫And I didn't for the longest time, Mongo DB, for example, didn't have Atomic City across documents.

155
00:11:11,360 --> 00:11:13,780
‫It has only atomos of the across a single document.

156
00:11:14,060 --> 00:11:20,690
‫And people are fine with that because most of the time they will do their data model so that it is a

157
00:11:20,690 --> 00:11:27,170
‫single document and instead of not, say, normalized like we do here.

158
00:11:28,940 --> 00:11:32,420
‫All right, let's jump into isolation.

159
00:11:32,570 --> 00:11:37,880
‫All right, guys, so I went ahead and inserted a few rows in the table that you don't have to see the

160
00:11:37,880 --> 00:11:38,280
‫process.

161
00:11:38,300 --> 00:11:39,230
‫I'm going to cut it out.

162
00:11:39,590 --> 00:11:41,190
‫It's a straightforward thing.

163
00:11:41,210 --> 00:11:43,970
‫We sold a couple of units it here, a couple of units right here.

164
00:11:44,000 --> 00:11:52,100
‫So now let's test out isolation and to decide isolation.

165
00:11:52,110 --> 00:11:54,110
‫Guys, I want to do the following.

166
00:11:54,120 --> 00:11:54,320
‫Right.

167
00:11:54,560 --> 00:12:02,330
‫So to this isolation, guys, I want to do the following just with normal transactions.

168
00:12:02,720 --> 00:12:06,710
‫I'm going to start a transaction in my first session right here.

169
00:12:07,070 --> 00:12:12,170
‫And I'm going to do a normal query saying select product and the count of the product.

170
00:12:13,430 --> 00:12:13,930
‫All right.

171
00:12:14,240 --> 00:12:22,130
‫I just want to see how how much sales did I made on getting our for each product from sales and group

172
00:12:22,130 --> 00:12:23,880
‫by group.

173
00:12:24,320 --> 00:12:31,510
‫If I do that all of a sudden now and this query, I have three sales for like two and three sales for

174
00:12:31,520 --> 00:12:34,250
‫product one, which is pretty good right now.

175
00:12:34,250 --> 00:12:42,620
‫I want to actually print the report that that shows that what are this list?

176
00:12:42,740 --> 00:12:43,700
‫Just the whole thing.

177
00:12:43,730 --> 00:12:47,900
‫So in order to do that, you do select product.

178
00:12:49,300 --> 00:12:52,330
‫Price, maybe you do a join, but we don't.

179
00:12:52,600 --> 00:12:57,160
‫We're not going to do that quantity from sales.

180
00:12:58,390 --> 00:12:58,730
‫Right.

181
00:12:59,140 --> 00:13:05,590
‫But before you going to execute that, you're going to execute at another session started right here.

182
00:13:06,810 --> 00:13:11,610
‫OK, and inserted a new sail, OK, you just inserted a new sail.

183
00:13:12,960 --> 00:13:20,100
‫Right, and you are just doing your own thing right here and generating this large report, but if you

184
00:13:20,100 --> 00:13:27,090
‫reach this stage and meanwhile think so, think of these two US as concurrent application transaction

185
00:13:27,270 --> 00:13:28,500
‫going at the same time.

186
00:13:28,740 --> 00:13:31,210
‫Someone is making a sale, someone is generating a report.

187
00:13:31,500 --> 00:13:34,200
‫So in the same time, I'm going to generate a new cell.

188
00:13:34,200 --> 00:13:37,700
‫Hey, we just sold a new phone, ram pedi.

189
00:13:37,990 --> 00:13:43,740
‫Analysts say price and quantity, home values.

190
00:13:44,160 --> 00:13:45,090
‫We sold the new iPhone.

191
00:13:47,550 --> 00:13:57,720
‫And at a price nine nine nine nine nine nine is just said, we sold 10 more phones and we also go ahead

192
00:13:57,720 --> 00:14:01,130
‫and update the products where sit.

193
00:14:02,130 --> 00:14:14,100
‫Quantity, quantity, no set inventory, equal inventory milestone where the equation because we just

194
00:14:14,850 --> 00:14:19,620
‫we just made etc. So we need to decrement and then what we did is we actually committed.

195
00:14:19,860 --> 00:14:20,310
‫We're done.

196
00:14:20,460 --> 00:14:21,120
‫We made us all.

197
00:14:22,460 --> 00:14:24,000
‫And we weren't back here.

198
00:14:24,060 --> 00:14:27,210
‫We're continuing to generate the report again, this is happening in parallel.

199
00:14:27,480 --> 00:14:29,310
‫I'm posing it just to show you.

200
00:14:29,700 --> 00:14:33,870
‫But this is now we made a sale and then you printed the record.

201
00:14:34,410 --> 00:14:35,580
‫So you printed that.

202
00:14:35,760 --> 00:14:39,270
‫You made three or three sales, three sales here.

203
00:14:39,600 --> 00:14:44,160
‫And then you made and then you actually printed that stuff.

204
00:14:44,160 --> 00:14:44,880
‫And guess what, guys?

205
00:14:45,400 --> 00:14:46,500
‫This don't match.

206
00:14:47,550 --> 00:14:50,160
‫You said that you and this happens a lot.

207
00:14:50,160 --> 00:14:50,460
‫Right.

208
00:14:50,460 --> 00:14:55,440
‫And actual companies, where were the actual table doesn't match the counts.

209
00:14:55,680 --> 00:14:59,240
‫And this happens if you do this kind of things.

210
00:14:59,290 --> 00:15:00,360
‫This I'm going to talk about.

211
00:15:00,360 --> 00:15:01,470
‫This is called isolation.

212
00:15:02,310 --> 00:15:02,840
‫Isolation.

213
00:15:02,850 --> 00:15:04,370
‫We're not fully isolated here.

214
00:15:05,220 --> 00:15:07,740
‫So when it will happen here, we printed this poppy.

215
00:15:07,740 --> 00:15:08,000
‫Right.

216
00:15:08,400 --> 00:15:18,780
‫And we said, OK, we made three phones, actually four phones and two and three ear buds.

217
00:15:18,780 --> 00:15:22,380
‫So the ear buds are correct, but the phones are not correct.

218
00:15:22,860 --> 00:15:26,490
‫You just produced an inconsistent.

219
00:15:29,160 --> 00:15:37,230
‫View, so inconsistency or consistency in general can happen because of animosity, which we explained

220
00:15:37,650 --> 00:15:42,270
‫and can also happen because of isolation, which is what I'm about to explain here.

221
00:15:42,570 --> 00:15:45,960
‫So how do I make sure that nobody.

222
00:15:47,370 --> 00:15:54,630
‫Touches my view when I am beginning my transaction, if I start in this transaction, I want you to

223
00:15:54,630 --> 00:16:02,580
‫give me a snapshot, right, of all the coins, and I'm going to do as if at the moment of the where

224
00:16:02,580 --> 00:16:03,660
‫the transaction started.

225
00:16:04,760 --> 00:16:08,600
‫And that's called a repeatable rate, so it's going to roll back this thing.

226
00:16:09,890 --> 00:16:18,330
‫And then we would begin a transaction, but we're going to do isolation level, repeatable rate.

227
00:16:18,350 --> 00:16:21,200
‫I think that's how you say the default was.

228
00:16:22,490 --> 00:16:24,620
‫Reid committed what that means.

229
00:16:25,790 --> 00:16:32,450
‫Anything that has been committed on other transaction, I, I can see it and this isolation level,

230
00:16:32,450 --> 00:16:38,280
‫Merks most of the time, sometimes in this case you don't want it, you don't want a repeat operator

231
00:16:38,290 --> 00:16:39,160
‫or serializable.

232
00:16:39,410 --> 00:16:41,870
‫So let's go ahead and start in a repeatable read.

233
00:16:42,260 --> 00:16:48,170
‫And now if I'm going to make my beautiful query, the count, hey, there are three and four, which

234
00:16:48,170 --> 00:16:49,060
‫is which is good.

235
00:16:49,070 --> 00:16:55,910
‫So we made four sales bad like that, but exclude the second query.

236
00:16:57,170 --> 00:17:02,510
‫Took away the sale, but in the meantime, someone started another transaction, we don't care what

237
00:17:02,600 --> 00:17:05,390
‫isolation level at that point where they.

238
00:17:06,620 --> 00:17:08,210
‫They inserted another.

239
00:17:09,650 --> 00:17:10,220
‫Sayle.

240
00:17:11,300 --> 00:17:12,140
‫Email, exactly.

241
00:17:13,730 --> 00:17:14,540
‫And they committed.

242
00:17:17,920 --> 00:17:20,030
‫So they just made it into the sale.

243
00:17:21,330 --> 00:17:21,960
‫What?

244
00:17:23,070 --> 00:17:24,270
‫If you query this.

245
00:17:25,420 --> 00:17:28,250
‫Notice that it didn't show up on your case here.

246
00:17:29,020 --> 00:17:29,350
‫Why?

247
00:17:29,350 --> 00:17:36,020
‫Because you're your isolation level prevents you from seeing stuff that other people's changing.

248
00:17:36,580 --> 00:17:38,500
‫That's that's a beautiful repeatably.

249
00:17:38,530 --> 00:17:39,430
‫It's expensive.

250
00:17:40,420 --> 00:17:45,160
‫You can understand how it works because it works at them like sea level.

251
00:17:45,280 --> 00:17:47,290
‫And Bosco's does it differently.

252
00:17:47,590 --> 00:17:49,000
‫Bicyclers and differently.

253
00:17:49,000 --> 00:17:50,560
‫SQL Server does it differently.

254
00:17:50,560 --> 00:17:54,840
‫Oracle does it exact very similar to bicycle bicycle.

255
00:17:54,850 --> 00:17:57,820
‫This is very similar to Oracle, if you want to be very precise.

256
00:17:58,540 --> 00:18:07,210
‫But that said, now, if you you have a consistent view, you made five sales technically on the iPhone.

257
00:18:07,210 --> 00:18:17,920
‫But if you query again that group by you only see four, you're in this other other query for other

258
00:18:18,130 --> 00:18:18,730
‫session.

259
00:18:18,820 --> 00:18:25,220
‫If you just select the ID count copied from what is that?

260
00:18:25,240 --> 00:18:27,490
‫So we have five units here.

261
00:18:27,580 --> 00:18:33,220
‫But here, no matter how many times you execute that, you're going to get a repeatable read your readers

262
00:18:33,220 --> 00:18:33,970
‫a repeatable.

263
00:18:33,970 --> 00:18:35,080
‫It's not going to change.

264
00:18:35,080 --> 00:18:39,460
‫And that's a very attractive feature for a lot of applications.

265
00:18:39,770 --> 00:18:40,010
‫Right.

266
00:18:40,510 --> 00:18:46,330
‫So now once you commit or roll back, you're done if you execute equity again.

267
00:18:48,050 --> 00:18:50,240
‫You see five because you're out of your transaction.

268
00:18:50,690 --> 00:18:55,250
‫All right, guys, that was isolation, finally for the fun of it.

269
00:18:56,620 --> 00:18:58,700
‫Let's do this, guys.

270
00:18:58,700 --> 00:19:01,820
‫Exit right here and then Dr..

271
00:19:02,830 --> 00:19:07,760
‫I'm going to do this, I'm going to begin a transaction, and then I'm going to insert a new product

272
00:19:07,780 --> 00:19:14,950
‫inserted to explain durability and that new product, and I'm going to prepare here to kill the content

273
00:19:14,960 --> 00:19:16,390
‫of the whole the whole container.

274
00:19:16,650 --> 00:19:17,570
‫You stop, right?

275
00:19:18,070 --> 00:19:27,130
‫But in the exact same second, I am going to insert a new sales and see if this is actually going to

276
00:19:27,130 --> 00:19:27,700
‫happen.

277
00:19:27,700 --> 00:19:31,760
‫If I say if it says commit, what will happen here.

278
00:19:32,380 --> 00:19:32,920
‫Name.

279
00:19:34,090 --> 00:19:41,080
‫Price inventory, yeah, that's just inserts a new product that's called.

280
00:19:42,180 --> 00:19:42,990
‫TV.

281
00:19:46,040 --> 00:19:53,120
‫Yeah, three thousand dollars and there is only 10 units of this, and I can do this and I'm going to

282
00:19:53,120 --> 00:19:57,370
‫commit, but at the same time we're going to kill the container.

283
00:19:59,340 --> 00:20:01,260
‫So immediately we were killed.

284
00:20:01,940 --> 00:20:02,570
‫So now.

285
00:20:03,750 --> 00:20:11,420
‫I want to know, did my TV get recorded that Postgres told us that it's committed, right?

286
00:20:12,690 --> 00:20:19,860
‫If I do this and a lot of their base doesn't support this, by the way, like this, but it is in the

287
00:20:19,860 --> 00:20:20,400
‫movie database.

288
00:20:20,400 --> 00:20:22,600
‫But it does support persistance, right?

289
00:20:25,440 --> 00:20:31,050
‫Essentially adorability is that means if you tell me that you committed something, that means if I

290
00:20:31,050 --> 00:20:34,970
‫log in again, I better be there to see if it is out there.

291
00:20:37,600 --> 00:20:44,230
‫Yup, it's right there because it told me it committed if if it didn't return the comment, that means

292
00:20:44,230 --> 00:20:49,810
‫it hasn't been written and durability is a very you might say you might think of it as like, OK, of

293
00:20:49,810 --> 00:20:52,960
‫course, Hosain, it's given that it has to be.

294
00:20:54,620 --> 00:21:01,070
‫If I write, something has to be there by noon, some is right to memory first for a performance reason

295
00:21:01,130 --> 00:21:07,370
‫and the container or the host dies, they forget to flash to disk.

296
00:21:07,700 --> 00:21:09,110
‫You just lost your stuff.

297
00:21:09,960 --> 00:21:12,100
‫But, guys, what do you think about this, guys?

298
00:21:12,640 --> 00:21:14,110
‫Thank you so much for watching.

299
00:21:14,870 --> 00:21:15,500
‫How fun.

300
00:21:15,530 --> 00:21:19,440
‫Enjoy the rest of the course that I'm going to see on the next one.

301
00:21:20,150 --> 00:21:20,810
‫Goodbye.