1 00:00:00,210 --> 00:00:03,810 Helen will come back more of the US practice questions. 2 00:00:03,900 --> 00:00:12,210 And here is another one why doesn't your support increasing the Ardis database instant storage for Microsoft 3 00:00:12,210 --> 00:00:13,100 sequel server. 4 00:00:13,280 --> 00:00:14,530 There is no one limitation. 5 00:00:14,580 --> 00:00:16,810 And because of which we cannot do that. 6 00:00:16,920 --> 00:00:19,380 How about the other ideas of the big engines. 7 00:00:19,380 --> 00:00:21,230 Can we increase the storage. 8 00:00:21,230 --> 00:00:22,130 Yes you can. 9 00:00:22,500 --> 00:00:24,780 So what are the answers. 10 00:00:24,780 --> 00:00:28,670 A Microsoft sequel server is not a native. 11 00:00:28,680 --> 00:00:30,140 It obviously there is engine. 12 00:00:30,150 --> 00:00:32,130 Thats why I would vote against it. 13 00:00:32,130 --> 00:00:32,500 Why. 14 00:00:32,520 --> 00:00:39,000 Because Oracle is not native us equal is not only native E.W. us but they you support it b because of 15 00:00:39,000 --> 00:00:45,240 the extensibility limitation of striped storage attached to the Microsoft Windows Server and that is 16 00:00:45,240 --> 00:00:49,730 that documented known reason See this is not correct. 17 00:00:49,740 --> 00:00:55,530 You can increase storage on Microsoft sequel server or this instance and this is wrong. 18 00:00:56,130 --> 00:00:57,740 There is no documented limitation. 19 00:00:57,750 --> 00:00:58,470 There is. 20 00:00:58,470 --> 00:01:00,770 So the correct answer is B. 21 00:01:00,990 --> 00:01:08,360 Next question your junior sit administrator is wondering if it is possible to downsize the RDA. 22 00:01:08,410 --> 00:01:09,980 Did it be instant storage. 23 00:01:10,080 --> 00:01:11,970 After it was launched. 24 00:01:12,290 --> 00:01:19,410 So you have launched an idea that is instance and lets say it has one terabyte of storage can you. 25 00:01:19,410 --> 00:01:28,940 After you launched it can you bring it down to let's say 500 gigabytes instead of one terabyte answer. 26 00:01:28,960 --> 00:01:31,200 The database engine is what determines that. 27 00:01:31,210 --> 00:01:33,020 And this is wrong. 28 00:01:34,090 --> 00:01:37,650 B It is possible for my secret but not for other engines. 29 00:01:37,660 --> 00:01:39,520 And again this is wrong. 30 00:01:40,090 --> 00:01:44,920 See it is not possible to scale storage down you can only scale it up. 31 00:01:44,920 --> 00:01:49,470 And this is the correct one so this is again and this is not a tweaked question or anything. 32 00:01:49,510 --> 00:01:56,230 This is something you need to know that it was already a service will not allow you to scale down or 33 00:01:56,230 --> 00:01:58,980 downsize the storage that you have. 34 00:01:59,020 --> 00:02:03,760 You can keep it as is when you create when you create the instance you can scale it up if you want. 35 00:02:03,760 --> 00:02:13,530 And if you need to but you cannot get it done the audience does not allow changing the size at all after 36 00:02:13,580 --> 00:02:14,910 that instance launch. 37 00:02:14,910 --> 00:02:16,640 This is partially true. 38 00:02:16,710 --> 00:02:19,590 You cannot change it down but you can change it up. 39 00:02:19,620 --> 00:02:22,700 So thats why I see it is the correct answer. 40 00:02:22,740 --> 00:02:27,670 Next question one should I choose provision IOPS over standard. 41 00:02:27,700 --> 00:02:35,690 Are the US storage so you have in the abs you know that the US use on the ABS does not use instant stores 42 00:02:35,700 --> 00:02:43,770 so if you have a question in the exam that challenges you which US which block storage that you can 43 00:02:43,770 --> 00:02:50,310 use with the R D s instances and one of the choices is instant store 100 percent it is not one of the 44 00:02:50,310 --> 00:02:51,190 options. 45 00:02:51,420 --> 00:02:57,330 So you need to focus on the remaining options of the of the answers or the multiple choices but the 46 00:02:57,330 --> 00:02:59,390 instant store is deaf and cannot support. 47 00:02:59,460 --> 00:03:04,850 The question is straight asking you can use instant store volumes with obvious instances. 48 00:03:04,910 --> 00:03:07,330 So the answer is an absolute no. 49 00:03:07,400 --> 00:03:08,620 It's non-supported. 50 00:03:09,360 --> 00:03:16,710 So Tony yes but you know the ABS has general purpose has the provision IOPS and has the magnetic among 51 00:03:16,890 --> 00:03:17,400 other types. 52 00:03:17,400 --> 00:03:18,330 Right. 53 00:03:18,360 --> 00:03:25,050 So here's the question is when should you prefer to choose provision IOPS over the standard audio storage 54 00:03:25,080 --> 00:03:28,140 over this one or this one. 55 00:03:29,190 --> 00:03:36,780 A for batch oriented workloads so batch processing be for workloads that are not sensitive to consistent 56 00:03:36,780 --> 00:03:40,370 performance which is I mean this one is wrong. 57 00:03:40,440 --> 00:03:41,190 This one is wrong. 58 00:03:41,190 --> 00:03:41,550 Why. 59 00:03:41,550 --> 00:03:45,920 Because provision IOPS will provide consistent performance. 60 00:03:45,960 --> 00:03:52,480 So why would you choose it because you don't need that consensus performance doesn't make sense. 61 00:03:52,550 --> 00:03:57,090 See for small workloads smaller to those you can live with the magnetic You can live with a general 62 00:03:57,090 --> 00:03:57,630 purpose. 63 00:03:57,630 --> 00:04:06,710 You don't need the provision types the for production online transaction processing all T-P workloads 64 00:04:06,750 --> 00:04:08,570 and this is the correct answer. 65 00:04:08,570 --> 00:04:16,710 That's the command when we use it for production online transaction processing workloads that require 66 00:04:16,710 --> 00:04:20,830 consistent performance high rate of IOPS. 67 00:04:21,120 --> 00:04:23,270 This is where you need the provision lives. 68 00:04:23,280 --> 00:04:30,630 So C is incorrect ABC are incorrect the is the correct answer and an important refresher for this one 69 00:04:30,630 --> 00:04:31,600 that you need to remember. 70 00:04:31,640 --> 00:04:35,610 Amazon RLDS provision of storage to improve performance. 71 00:04:35,610 --> 00:04:44,100 So use the Amazon Diest provision IOPS to improve performance for any production application that requires 72 00:04:44,310 --> 00:04:52,650 fast and consistent performance with a common provision provision of storage type that delivers fast 73 00:04:52,740 --> 00:04:56,280 predictable and consistent throughput performance. 74 00:04:57,150 --> 00:05:05,080 And it is optimized for online transaction processing workloads that have consistent performance requirements. 75 00:05:05,160 --> 00:05:06,360 So we have workloads. 76 00:05:06,360 --> 00:05:13,680 If the question was different and they said you have an application that requires consistent database 77 00:05:13,770 --> 00:05:19,800 read and write performance and this is a must have requirement what would you recommend as a storage 78 00:05:19,800 --> 00:05:21,790 option for this instance. 79 00:05:21,810 --> 00:05:28,740 And again it becomes the audience if the question asks you that you have an OLTP workload and you are 80 00:05:28,740 --> 00:05:34,530 about to architect a solution that has already passed instances which is volume which storage would 81 00:05:34,530 --> 00:05:40,020 you choose and then Ill give you the magnetic The general purpose there and provision IOPS which one 82 00:05:40,020 --> 00:05:40,560 would you choose. 83 00:05:40,560 --> 00:05:40,990 Okay. 84 00:05:41,070 --> 00:05:47,100 So that same question can be tweaked in different ways but you need to be aware that production workloads 85 00:05:47,100 --> 00:05:55,080 consistent performance and oil T.P. kind of workloads can predictable and fast performance. 86 00:05:55,080 --> 00:05:57,780 All of that is exactly is provisioned. 87 00:05:57,900 --> 00:06:05,250 The next question is it possible for clients to test there are database instances against a new version 88 00:06:05,250 --> 00:06:06,800 before deciding to upgrade. 89 00:06:06,990 --> 00:06:12,880 So now I am at version six point zero for instance of my and my sequel and I would like to go to 6.5 90 00:06:13,440 --> 00:06:18,410 but I have a production and the production is very important to me and I don't want to upgrade. 91 00:06:18,540 --> 00:06:25,140 And then I start finding issues in 6.5 and I have to roll back again to one point to 6 percent. 92 00:06:25,650 --> 00:06:34,590 So is there a way I can somehow test 6.5 and then if I am comfortable that this is what I want and will 93 00:06:34,590 --> 00:06:40,650 deliver they expected then I would go to my production and take that from six point thirty six point 94 00:06:40,650 --> 00:06:41,010 five. 95 00:06:41,010 --> 00:06:42,730 Thats where the question is asking. 96 00:06:42,870 --> 00:06:47,950 In this instance the answer is no. 97 00:06:48,020 --> 00:06:52,710 It is our this is full is a fully managed service and you just need to put a request for it to be asked 98 00:06:52,740 --> 00:06:55,810 to upgrade does that make sense if you think about it doesn't make sense. 99 00:06:55,810 --> 00:07:01,470 So you put in a request and then your old production environed becomes messy because of a bug or because 100 00:07:01,470 --> 00:07:06,910 of inconsistency or because the new version does not deliver what you are expected be. 101 00:07:07,140 --> 00:07:08,210 Yes it is possible. 102 00:07:08,280 --> 00:07:15,900 Create a snapshot off of the existing Ardis instance create a new artist instance of the snapshot then 103 00:07:15,900 --> 00:07:18,770 upgrade the new instance then do testing. 104 00:07:18,780 --> 00:07:24,010 So I have lets say its multis you have my primary production my standby. 105 00:07:24,330 --> 00:07:25,670 So what this is suggesting. 106 00:07:25,680 --> 00:07:27,590 Why dont you go ahead and. 107 00:07:27,840 --> 00:07:28,890 Exactly. 108 00:07:28,890 --> 00:07:32,060 So stand by standby create a snapshot of the snapshot. 109 00:07:32,070 --> 00:07:39,670 Create another Ardis instance that has nothing to do with the primary and stand by the production ones 110 00:07:41,100 --> 00:07:49,210 and of the not when you have the earliest instance in should the request to AWOS to upgrade to 6.5. 111 00:07:49,520 --> 00:07:54,340 OK so they upgrade the instance to 6.5 same instance. 112 00:07:54,480 --> 00:08:00,980 Now do all your testing and and simulated production testing on this instance. 113 00:08:00,990 --> 00:08:06,780 If everything looks fine then you can go ahead and submit a request to upgrade this one also to 6.5 114 00:08:07,290 --> 00:08:08,280 and you can choose. 115 00:08:08,280 --> 00:08:13,470 By the way when you when you want to do any changes to your ideas did it be like you want to upgrade 116 00:08:13,470 --> 00:08:15,300 you want to scale up and all that stuff. 117 00:08:15,300 --> 00:08:20,790 You can choose for that change to happen immediately or you can choose to wait until the maintenance 118 00:08:20,790 --> 00:08:23,410 window when this will will happen. 119 00:08:23,430 --> 00:08:24,720 So you have the choice. 120 00:08:25,060 --> 00:08:30,690 OK so this one looks as feasible and logical doable option. 121 00:08:31,170 --> 00:08:32,570 But lets go through CND. 122 00:08:32,790 --> 00:08:38,190 You need to launch an easy to instance and install the new database engine version test then upgrade 123 00:08:38,190 --> 00:08:38,790 your data. 124 00:08:38,820 --> 00:08:45,890 Are these instances this is not correct because you need to do that on an already managed instance instead. 125 00:08:45,930 --> 00:08:49,500 Instead there might be commonalities but is it 100 percent. 126 00:08:49,500 --> 00:08:55,260 No you need to simulate the same environment on a test bed and then you find out you're testing whether 127 00:08:55,260 --> 00:08:56,550 it passes everything. 128 00:08:56,580 --> 00:08:58,450 And then you can upgrade it. 129 00:08:58,470 --> 00:08:59,790 So this one is wrong. 130 00:09:00,300 --> 00:09:01,430 It is very close. 131 00:09:01,570 --> 00:09:03,960 The No you cannot test before upgrade. 132 00:09:03,960 --> 00:09:09,720 You can test after a grade which doesn't make sense because as I mentioned before if you upgrade and 133 00:09:09,720 --> 00:09:14,690 then you have problems then you have to roll back and you cannot do testing after you upgrade. 134 00:09:14,940 --> 00:09:15,360 Why. 135 00:09:15,360 --> 00:09:21,060 Because we know that it is only 30 minutes for the maintenance window that you have and doing that 30 136 00:09:21,060 --> 00:09:27,060 minutes you cannot do a lot of testing and all that so you are better off doing this and then you can 137 00:09:27,450 --> 00:09:27,870 upgrade 138 00:09:30,940 --> 00:09:37,750 because so a C and D are wrong and B is the correct answer and a quick refresher or two that you can 139 00:09:37,750 --> 00:09:42,150 test a database engine against a new version before upgrading. 140 00:09:42,520 --> 00:09:51,340 You do this by creating a database up shot of your existing data is instance and if it is not easy this 141 00:09:51,340 --> 00:09:54,090 will be created from the standby to avoid suspending. 142 00:09:54,090 --> 00:10:00,880 I was on the primary second step restoring from the database snapshot to create a new database instance 143 00:10:02,350 --> 00:10:05,900 initiating a version upgrade for the new database instance. 144 00:10:05,950 --> 00:10:11,030 And now you have a new version test that you can test and then you can do your test. 145 00:10:11,030 --> 00:10:16,540 You can experiment safely on the upgraded clone of your data instance before deciding whether or not 146 00:10:16,540 --> 00:10:20,590 to upgrade your original production database instance 147 00:10:24,240 --> 00:10:31,350 another question you asked our guest if not otherwise specified will perform Ardelia's instance changes 148 00:10:31,490 --> 00:10:37,880 using the maintenance window the user has the option to select a maintenance window. 149 00:10:38,040 --> 00:10:38,850 So that's a fact. 150 00:10:38,910 --> 00:10:45,000 Now we're sitting in fact now if you launched an arduous instance and did not specify the maintenance 151 00:10:45,000 --> 00:10:46,600 wounded during launch. 152 00:10:46,920 --> 00:10:53,490 When will it be as ready as carry out the required maintenance or implement changes the user requests 153 00:10:53,610 --> 00:10:55,540 to be done. 154 00:10:55,560 --> 00:10:57,160 So what is what. 155 00:10:57,200 --> 00:11:04,770 The question is simply you launched an obvious instance during the launch you have the option to check 156 00:11:04,770 --> 00:11:09,170 to select when your maintenance window is going to be for this or this instance. 157 00:11:09,750 --> 00:11:12,840 You skipped that and you launch the obvious instance. 158 00:11:12,840 --> 00:11:19,920 The question now is if you want to do any changes here and you don't want that to be immediate What 159 00:11:19,920 --> 00:11:21,230 will our ideas do. 160 00:11:21,240 --> 00:11:29,400 Given that you have not specified a maintenance when did you launch a will not allow any updates or 161 00:11:29,400 --> 00:11:32,410 changes until the user configures a maintenance window. 162 00:11:32,460 --> 00:11:34,330 And this is wrong. 163 00:11:34,900 --> 00:11:41,340 B If the user requests any changes or updates AWOS will prompt the user for a maintenance window. 164 00:11:41,700 --> 00:11:43,800 And again this is not how it works. 165 00:11:44,100 --> 00:11:50,730 See the user must terminate the previous instance and launch a new one and specify during the launch 166 00:11:50,730 --> 00:11:53,980 of the new one the desired maintenance window. 167 00:11:54,530 --> 00:12:00,270 Frank this will select the default maintenance window if not specified by the user and that's the one 168 00:12:00,270 --> 00:12:02,320 that makes the most sense. 169 00:12:02,370 --> 00:12:07,430 Why do you need to rip apart your are instance just because of a change window. 170 00:12:09,290 --> 00:12:10,610 And it's not going to prom. 171 00:12:10,610 --> 00:12:15,620 Q When you do the change it will ignore the fact that he did not specify. 172 00:12:15,650 --> 00:12:20,690 I mean a maintenance window and there is a default depending on the region depending on where you launch 173 00:12:20,690 --> 00:12:29,240 that instance there will be a default maintenance window that it will pick it up for you and will do 174 00:12:29,240 --> 00:12:30,670 your changes during that. 175 00:12:30,680 --> 00:12:33,000 And that's the one that makes the most sense. 176 00:12:33,500 --> 00:12:38,070 So it is incorrect because incorrect is incorrect and the is the correct answer. 177 00:12:38,210 --> 00:12:45,590 Refresher every instance has a weekly maintenance window and that could be also a question maybe more 178 00:12:45,590 --> 00:12:49,940 to the office but you never know maybe charge up the maintenance window for you. 179 00:12:49,940 --> 00:12:59,390 Are these instances that you launch in any region in a VPC is weekly daily hourly monthly or yearly. 180 00:12:59,390 --> 00:13:01,060 Which one is the right one. 181 00:13:01,100 --> 00:13:01,510 Exactly. 182 00:13:01,510 --> 00:13:03,180 It's a weekly maintenance window. 183 00:13:04,070 --> 00:13:10,670 If you did not specify one at the time you create that it would be instance it will choose one randomly 184 00:13:10,670 --> 00:13:13,050 for you and that's 30 minutes long. 185 00:13:13,050 --> 00:13:13,730 1. 186 00:13:13,760 --> 00:13:17,120 All right so let's take a break and we'll continue on. 187 00:13:17,120 --> 00:13:18,500 Don't go anywhere please. 188 00:13:18,500 --> 00:13:20,520 We'll continue on with more. 189 00:13:20,710 --> 00:13:21,440 Yes. 190 00:13:21,590 --> 00:13:24,420 Practice questions because practice makes perfect. 191 00:13:24,590 --> 00:13:25,550 And I'll see you in the next election.