Google Compute - Optimization Costs and Perfomance in Google Cloud Platform

1. Step 01 - Understanding Sustained Use Discounts in GCP - Google Cloud Platform

Welcome back. Until now, we have been playing with virtual machines. Starting with this step, let's look at a few important things that you need to understand about keeping your costs low. Whenever you are making use of the cloud, you want to keep your costs as low as possible. What are the things that you can do to keep the cost of your virtual machines low? The first thing that we will be talking about is sustained use discounts. What is a sustained use discount? It's an automatic discount which is applied for running VM instances for a significant portion of the billing month. For example, if you are using one or two machine types for more than 25% of the time period of the month, then you get a 20% to 50% discount on every incremental minute. So beyond that, whatever time period you are running for, you'd get a discount on it,and the discount increases with usage. The amount of discount you would get would be higher than the amount of discount you would get if you were running it for 40%. And the great thing about the sustained use discounts is that you don't really need to do anything. There is nothing you need to do specifically for this. When you are creating or running your virtual machines,this is automatically applied by the platform itself. However, there are a few important restrictions that you need to remember a little later in the course. We will also be talking about Google KubernetesEngine, which is a managed Kubernetes service which is provided by the Google Cloud platform. Whenever you are creating instances using the Google Compute Engine, that's basically what you are using in the course. Until now, or whenever you create instances with the Google Kubernetes engine, the sustained use discounts have been applied automatically. Remember that this does not apply to certain machine types. For example, for e two and a two, the sustained use discounts do not apply to them. And also remember that if you're using AppEngine, flexible or dataflow to create the VMs,sustained use discounts do not apply. We'll talk about the App Engine and data flow as we go further in this course. The important thing to remember about sustained discounts is the fact that these are automaticallyapplied when you use instances that are created by Google Kubernetes Engine and Compute Engine for a significant part of a specific month. Also, remember that there are a few restrictions. I'm sure you're having an interesting time, and I'll see you on the next step.

2. Step 02 - Understanding Committed Use Discounts in GCP - Google Cloud Platform

Welcome back. In this step, let's look at another type of discount that is applied to your virtual machines. How would this type of discount need a commitment from you? And that's why they are called Committed Use Discounts. Using Committed Use Discounts is recommended for workloads with predictable resource needs. You can say I would need a virtual machine of this type for one year or three years, and depending on the machine type and the GPUs you are asking for, you can get up to a discount. The discount that you get with Committed UseDiscount is higher than the discount that you would get with Sustained Use Discounts. So if you're sure that you need a specific virtual machine for one year or three years, then it makes sense to go ahead and register for a Committed Use Discount because this gives you more discount than depending on a Sustained Use discount. Again, the committed use discount is applicable to instances which are created by the Google Kubernetes engine and the Compute engine. And the same restriction that we talked about when we talked about Sustained Use discounts applies to Committed Use Discounts as well. This does not apply for VMs created by App Engine Flexible and Dataflow. Now, How can you ask for a committed use discount? Let's go back to our virtual machines. In the menu for virtual machines, the last option you will see is Committed Used Discounts. This is where you can actually go in and ask for a Committed Use Discount.This is where you can actually go in and purchase a commitment. So I can go ahead and say "Purchase a commitment" and "I want to purchase a specific commitment." As you can see here, you can actually purchase a hardware or software commitment. What we are interested in is a virtual machine commitment, which is a hardware commitment. So you can give it a name, you can choose which region you'd want to have the commitment in, and you can choose the specific family you'd want to commit to. Then you can specify how much time you want to commit to. You can also specify a little bit about the configuration of the machine you'd want. Do you want to coast? Do you want a memory? Do you want GPUs? Do you want local SSDs? And then you can go ahead and click "Purchase." I'm not going to click "Purchase" here because I don't really want to commit to it. What I would do is I would go ahead and say "Cancel." Once you register a commitment, if you create any new instances of that type,the discount is automatically applied. Instead, we talk about committed use discounts. These are recommended for workloads with predictable resource needs. You knew ahead of time that I would want to run a web application for a long time. In those kinds of situations, you can get the best discounts by going for a Committed Use Discount. I'll see you in the next step.

3. Step 03 - Saving Costs with Preemptible VMs

Welcome back into the fold. Let's talk about preemptible virtual machines. What are preemptible virtual machines? Preemptive virtual machines are short-lived, cheaper compute instances. If you are aware of AWS, then preemptibleVMs are very very similar to spot instances. The preemptible instances can be stopped by GCP at any time within 24 hours. The maximum time that you can run a preemptible instance is typically 24 hours. And you need to remember that GCP can terminate this instance at anytime within the 24 hour period as well. Just before GCP terminates your instance, you will get a 32nd warning. You can use this time period to save anything you might want to save. Now when do you go for a preemptible VM? You would use a preemptible VM if your applications are fault tolerant, you can stop and start them at any point in time, and you are very sensitive. You'd want to run the workload with minimum cost, and your workload is not immediate. If your users are accessing a specific application, then your workload is immediate. You'd want to do that immediately. However, if you have badge programmes which you can run, let's say within the next 15 days or within the next month, those are good workloads where you can go for preemptible VMs. So a good example is non-immediate batch processing jobs. Now let's look at a few restrictions. Preemptable VMs might not always be available. Number two, there is no SLA that is guaranteed by the Google cloud platform and your preemptibleVMs cannot be migrated to a regular VM. You cannot convert a preemptible VM into a regular VM. And number three is that you cannot automatically restart them if they have stopped you for any reason. You cannot actually restart them again. And the last but very important one as far as this course is concerned is that free tyre credits are not applicable for launching preemptible VMs. So I would recommend that you not launch preemptible VMs at all. Now let's try and see how you can actually launch a preemptible VM. We're not going to launch it, we're just going to see where you can actually launch it. So I'm going to VM instances and you can go and say create and this is the typical configuration we do. Where can I get a preemptible instance? The place where you can configure isin management, security, networking, and sole tenancy. If you go further down, if you go to the availability policy, that is where you can actually configure the preemptive VM. A preemptible VM costs much less but lasts only 24 hours. It may be terminated sooner due to system demands. And this is where you can actually set it to turn on. If you set preemptible to on and launch an instance, then it becomes a preemptible instance. In a step, we looked at preemptible VMs. These are short-lived, very cheapcompute instances which you can use to run non-immediate fault-tolerant workloads. If you have a non-immediate batch programme that you'd want to run at a very, very low cost, then you can opt for a preemptible VM. I'm sure you're having a wonderful time, and I'll see you in the next step.

4. Step 04 - Understanding Billing for Google Compute Engine - GCE VMs

Welcome back. In the previous steps, we discussed a few discounts that are available for your Google Computers engine. In this step, let's look at a few more important things around the building of your compute engine. It is very important to remember that you have built your compute engine by the second. However, this is only after a minimum of 1 minute. So if you start a new VM instance, you will pay for a minimum of 1 minute. Thereafter, you'd be billed by the second. Whenever you stop a compute engine, for instance,you are not billed for compute. However, remember that if you have any storage that is attached to the computer instance, you'd still be billed for it. The first recommendation when it comes to reducing your bill is to always create budget alerts. How can you create budget alerts? If you go over to the navigation menu on the left hand side, you should see something called "billing." In billing, you should see budgets and alerts. If you click that, you'd be landing on a specific screen where you can actually create a budget. Avoid surprises on your bill by creating budgets to monitor all your Google Cloud charges in one place. After you set a budget, you can create budget alerts to email, billing admins, and users when charges exceed a certain amount. So go ahead and create a budget. Let's create a budget right now. I would call this budget my first budget, and I'll choose the defaults that are suggested for me. So I'll choose all projects, all services, and I'll say next. And over here I can specify a target amount. I'll say the target amount for me is let's keep it as low as possible. So let's just say I would say "target budget is Rs 10 and I would go ahead and click Next. And if you click Next, you will see the alert threshold rules. So you can configure alerts at different thresholds. You can see that the defaults which are suggested are getan alert when 50% of the budget is utilized, which isRS5, get an alert when 90% of the budget is utilized, and also when 90% of the budget is utilized. You can further customise this threshold if you want, and if you scroll down,you can configure the notifications as well. So make sure that you check this out. So, email alerts to billing admins and users. This would ensure that you get an email when the cost exceeds your limits, and at the end of that, you can go ahead and click Finish. Now the other thing that you can do, in addition to creating a budget alert, is a bill export. So you can actually take the billing and export it to a number of services. A little later, we'll be talking about BigQuery,which is one of the big data services offered by Google Cloud. Using billing export, you can actually send your billing data to the BigQuery data set and, using BigQuery, you can run your queries as well. We'll talk a little bit more about billingexports when we talk about billing in depth a little later in the course. So, the best practise as far as managing your costs is concerned is to create budget alerts and also to make use of budget exports so that you are on top of billing all the time. We already looked at a few ways that you can save money with virtual machines Choose the right machine type and the right image for your workload and also be aware of the discounts available. So be aware of the sustained use discounts, and if you want, you can actually register for a committed use discount and also understand that you can get discounts by using preemptible virtual machine instances in a SEC. We talked a little bit more about the costs involved with your virtual machines created using your Google Compute engine. I'll see you in the next few days.

5. Step 05 - Achieving High Availability with Live Migration and Automatic Restart

Welcome back. In a step, let's look at another important feature of a compute engine, which is live migration. How do you keep your VM instances running when a whole system needs to be updated? Let's say GCP wants to perform a software or hardware upgrade on your specific virtual machine. You don't want the application to go down when the software or hardware upgrade is being performed to ensure that Compute Engine provides a feature called Live Migration. What happens with live migration is that your running instance is automatically migrated to another host in the same zone. So, if a software and hardware update is performed on a specific host, all running instances of that host will be migrated to another host in the same zone. And this does not change any of the properties of the VM. Live migration is also supported, for instance, even if you have local SSDs. However, it is important to remember that live migration is not supported for preemptible instances and instances to which you have added graphic processing units. Now, how do you configure live migration? How do you configure if you'd want to use live migration or not for your specific instance? The place you would want to configure this is the Availability Policy. This is very important to remember as far as the examination goes. The Availability Policy is where you would configure the live migration configuration. And as part of the policy,you can configure two important things. One is on Host Maintenance. What should happen during periodic infrastructure maintenance? The default is to migrate, which is to migrate your instances to another host on other hardware. The other thing you can also do is terminate. If you don't want your instance to be migrated,then you can choose Host Maintenance to terminate. Then the VM instance would be automatically stopped at the time of a migration. The other option you can also configure is automatic restart in the Availability Policy. You can also configure your VM instances to be automatically restarted if they are terminated due to any non user initiated reasons.So if you're not terminating the instance due to a maintenance event or due to a hardware failure, the instance will be terminated. In these cases, you can also specify whether or not the instance should be restarted automatically. This configuration can also be made in the Availability Policy. Let's see where the configuration is in the console right now. This is a screen from which you can create a VM instance. You can go in here. We already looked at the Availability Policy configuration when we were talking about preemptible instances. It's the same place we need to go right now as well. So I'll scroll down and go to Management Security Disks, and if you actually scroll down a little bit and go towards Availability Policy, This is where you can see the configuration for host maintenance and automatic restart. This is just below the preemptive t configuration. So, on host maintenance. When Compute Engine performs periodic infrastructure maintenance, it can migrate your VM instance to other hardware without downtime. And you can see that the default option, which is recommended, is to migrate the VM instance. However, if you want, you can actually choose to terminate the VM instance as well. And in addition, you can also consider automatic restart. Compute Engine can automatically restart VM instances if they are terminated for non-user-initiated reasons such as maintenance events, hardware failures, software failures, or other such reasons. And you can leave the automatic restart on, which is the recommended option and the default option as well. In a step, we looked at the modern option of having high availability for your virtual machines. You can set up host maintenance to migrate and you can enable automatic restart. I'm sure you're having a wonderful time, and I'll see you in the next step.

ExamSnap's Google Associate Cloud Engineer Practice Test Questions and Exam Dumps, study guide, and video training course are complicated in premium bundle. The Exam Updated are monitored by Industry Leading IT Trainers with over 15 years of experience, Google Associate Cloud Engineer Exam Dumps and Practice Test Questions cover all the Exam Objectives to make sure you pass your exam easily.