Google Compute Engine is a service-based infrastructure that allows clients to easily work through Google’s physical hardware and Google compute cloud. The Google computer engine exchanges a large number of scalable virtual machines to serve as clusters used for that purpose.
Contents in The Article
GCE can be managed through a RESTful API, command line interface, or web console. The computing engine is serviced for a minimum of 10-minutes per use. There is no up or front fee or time commitment. GCE competes with Amazon’s Elastic Compute Cloud (EC2) and Microsoft Azure.
Google compute engine?
Everyone i told you about the Google compute engine, what is it? what are the features and components. what can we do with it and I’ll show you how in just a few minutes we can spin up our own virtual machine in the Google compute engine.
We’ll throw a web server on it we’ll put a web page inside of that web server and serve it up to the world so Google compute engine is their infrastructure as a service offering, it’s one of the products in the Google cloud platform, that allows us to spin up virtual machines notice instances in the cloud hosted managed and running inside of Google’s data centers.
When we create an instance we’re going to specify a machine type ,which is going to determine the physical specification of that virtual machine the number of virtual CPUs the amount of memory.
How many persistent disks google compute cloud?
It can support that kind of thing persistent disks google compute cloud Google Compute Engine.
There’s a lot of tears in here to choose from and within these there are more tears. that will really allow you to customize your machine for the kind of workload that you’re expecting from there we have resources resources.
We’re going to create attach to our virtual machines and the real beauty behind google compute engine in Google’s cloud is that all of these resources are scoped meaning.
They’re independent so it’s a very modular system, if we were to create an instance and entire disk to it. if that instance were to get deleted or something bad were to happen in order to crash our resources are fine,
Because they’re independent we could easily spin up another machine reattach. the resources and be right back where we were and to go over some of these here.Google Compute Engine related.
We have disks known as persistent disks anytime you spin up an instance we needed this that contains an operating system in a root file system and we can attach more persistent disks that contain our data images provide everything that our instance needs to run so a boot loader and operating system and a root file system at the free level.
We have Debian and Sint OS at the premium level we have Red Hat Enterprise Linux souza and Windows Server 2008 r2. you can also create your own custom images from these flavors we also have networks and firewalls that we can create to control traffic to and from our instances addresses each (virtual machine) Google Compute Engine can have up to two addresses.
One internal required address for communicating within the network and an optional external IP address we also have regions and zones and this is where we can determine where our virtual machines (Google Compute Engine) and our data lives and a lot of these resources are scoped at either the zone level or the global level available to everything.
In fact disks are scoped at the zone , this is what makes them independent so they’re not tied to an instance and your data is replicated across the data centers in that zone. so it’s it’s a redundant and reliable instances are also scoped at the zone level.
So that means if we create a disk in a zone only instances within that zone will be able to use that disk addresses are scoped. At the region level images networks and firewalls are all known as global resources accessible to any zone within a project think of a project as your own private network. Your own cluster of virtual machines on the future front google compute engine.
Future front google compute engine?
Offers a pretty impressive server-side load balancing function that allows us to redirect incoming traffic across pools of (virtual machines) Google Compute Engine. so we can use this to distribute heavy load. we can use it as a way to provide redundancy when failures happen.
It’s really nice and we have full control over, because we can create forwarding rules that match and direct certain kinds of traffic to the load balancer that will then distribute It across these pools of machines.
I should also note that Google’s load balancer doesn’t require any pre-warming, so i can instantly scale to support those spiky traffic patterns. we also have sub our billing many cloud competitors out there do hourly billing your billed by the hour in Google’s cloud (Google Compute Engine).
Your billed by the minute with a minimum of 10 minutes, so if you use your instances for 12 minutes, you’re only paying for 12 minutes rather than a full hour and Google also.
Recently announced 30 to 85 percent price cuts across the Google cloud platform as well as sustained usage discounts. so the more you use it the bigger the discount global scope damages and this is a really cool feature.
Because if you’ve used any of the other clouds out there you know that you either can’t use images across regions or you need to copy your images across regions Google maintains a global repository for all images and custom images.
So it’s one spot and we can use them from anywhere, it also has a very robust and integrated networking system. you can create virtual networks.
private networks public subnets firewalls routers gateways ACLs the whole ball of wax and it’s all integrated and easy to work with you can’t really talk about Google’s cloud platform without mentioning performance and in the google compute engine.
It’s no different blazing fast network speed along with consistent disk i/o and as we saw when we talked about images. we have also Mortgage Board of Linux based operating systems available and Windows was just recently announced another brand new feature.
Recently announced I have migration and this is a big differentiator between google compute engine and the competition out there in the competition what happens, when your virtual machine needs to get taken down for maintenance, you’ll get an email a few weeks or a few days before that happens in compute engine.it’s all transparent it’s all under the hood
Google will actually migrate you instance to another instance, so they can take down the original instance for hardware or software maintenance and this is all transparent, you’ll never see it your instance could go down ten times in a day and you’ll never notice the difference so it all equates to zero downtime which is a pretty big deal out there in the IT industry these days and the fact that it’s transparent is just icing on the cake. so with all that said..
Google compute cloud
Why don’t I give you a quick demo to show you, how easy and quick it is to spin up a virtual machine in Google’s cloud. all right so the first thing we’re going to do is look at this from the easy perspective and that is using Google’s cloud console (Google Compute Engine). it’s got they have a nice GUI that you can use to do this stuff manually.
So next finish type stuff and then I’ll show you the really powerful way to do it and that is using the Google Cloud SDK there’s a couple of utilities in there G cloud and GC util that we can use to do all this stuff locally.
so it opens it up to scripting out all these actions so let’s bring up Chrome and I’ve already got a session ready to go here but if you want to get to this point head to cloud.google.com sign up and then create a project.
So let’s choose Google compute engine and then I’ll take us right into instances and here’s a bunch of resources here that we can work with so here’s disks here, if you want to see all the available images here.
They are they’re constantly keeping these distros updated as well here’s the latest Debian 7 for instance that. they just released today as a matter of fact and then they deprecated old ones.
But if we go back up to instances here and choose new instance fill out an instance name description meta data about it and then you can choose your zone machine type boot source image any additional persistent disk, that you want to attach and you can configure networking you hit the create button in literally seconds.
You will have an instance up and running that you can SSH into and do whatever you want with all right.
so let’s do this for real the hard core way using the Google Cloud SDK. so let’s bring up the terminal shell and this is the exact walkthrough that’s on Google’s into Google’s documentation for google compute engine.
So I’ll give you link for that if you want to try this on your own but essentially were to do here is first we need to download the Google Cloud SDK.
Start your free week https://cbt.gg/2LZhF9F
So we’re going to curl for it here inside of our terminal this will get everything all the all the binaries everything. we need here to work with Google Cloud locally and we’re just say yeah extract that here yeah overwrite it since I’ve already got it in there and then we will improve the customer experience sure.