- Redundancy
In this video, we’re going to be talking about redundancy and different things that we can add to our networks in order to make them redundant. In other words, if something fails, something else can kick in. We already talked about raid, which is really important in case a hard drive fails. But there are other things that we can implement in our network, such as having redundant power in case our main power line goes out. I do have a few things here that our exam wants us to know.
Let’s take a look. Okay, so the first thing we’re going to take a look at is this term called geographic dispersal. And what that means is that you basically want to spread your data not just in one geographic region but across multiple regions. Let me explain. If you have your data in one datacenter, in one part of the world, say the northeastern region of the United States, and there is a major blackout or a major flood in that region, the data center may be down for an extended period of time. If that data is only located in that particular geographic region, then you may not have access to it.
So what you should do is spread it out. For example, when designing redundant web servers, never load balance one without also load balancing the others. If you have a redundant server and one fails, the other one kicks in. This is called a cluster. If you have these types of clusters, you don’t want two web servers in the same region. You may want one to be in New York, and you don’t want one to be in California, which is a pretty big span there. It’s very weird, unless the apocalypse is coming, that New York is going to get hit with the same kind of natural disasters as California would at the same time. So you basically want to split your data geographically speaking. The other thing here we’ll talk about is load balancers. So load balancers are devices on a network that basically balance loads.
So I want to show you guys a load balancer. One well-known example is Netscape Citrix Netscape Enterprise Class Load Balancing Device. This is what it looks like. And notice it has some switchboards on it that basically disconnect service to it. And this balances the load across multiple servers. As an example, suppose you have the Internet, and then you connect the load balancer to the Internet, and the load balancer connects to all of the web servers. Web server. Web server. So as traffic is coming off the Internet, maybe because they want the pages on the web server, it’s going to distribute the load. And if there’s any kind of problem, that will determine that this device is no good and stop sending loads there.
Okay? So that’s what a load balancer is. Now, load balancers are really expensive devices that big organizations use. As you can see, this one is on sale for $50,000. As you can see, organizations will use this, especially when running big websites. If you’re running a large website, you can’t host it on just one computer. You’re going to want to distribute this across many different systems.
The other thing I have here is known as “NiCad teaming.” Now, NICAR team is something you mostly do on a Windows server; it used to be possible on Windows workstations, but not anymore. Essentially, this combines your NiCad into a single card that you can use. Let’s say you have two gigabit cards. It’s in there. It will allow you to send and receive data using both one gigabit of NIC. If one goes down, the other one keeps working. It’s a great option to do something that you should be doing, especially on servers. Okay, the other thing we have here is power. Now, power is really important. Blackouts are devastating to organizations. And if you don’t have this additional power, then you could be in a whole lot of trouble. For example, when your main light goes out, you may have a generator or an UPS. So upset.
There are a couple of things that we should know about UPS. So I have a picture here, somewhere along these lines. Where am I? Here. Where’s my UPS? Here we go. So here is an update from Amazon. I have exactly one of these things. Not the exact one, but a smaller one that I paid about $200 for, hooked up over there. It pulls up the sand. We hope you have a Nas hooked up to it. We’ve got a router hooked up to it. What this does is ensure that if we ever lose power, the UPS will kick in and be able to sustain our equipment for a little over an hour.
So this is just additional power. Now, there are some things we should know about these particular UPS. First of all, this will upset our battery base, which means over time the battery will die. For example, when we bought this UPS, we had about an hour and 20 minutes. It’s been about one and a half years. Now. The UPS has dropped all the way to about an hour. I think I went three minutes the last time I looked at it. It’s like all rechargeable batteries; the cell dies over time, right? Lithium-ion batteries both charge and discharge. Charge and they die. Now, the thing with UPS is that they run on batteries, and the computers are plugged into them, always running off of this battery.
And then the UPS plugs into the wall. So if there are any power outages, it doesn’t affect the machines because they’ve always been running on this battery. Computers don’t even know. But remember, upset doesn’t last long. Remember, it’s a battery. And the battery has to be pretty large, able to support long-term power. Ups are generally rated, or should be rated, for as long as it takes to turn your computers off, which is not long-term power. If you’re looking for long-term power, what you’re going to have to get is a generator. Generators are what you’re going to use in order to get power for days or weeks. Generators will need fewer supplies. Here is a generator. And notice that this one is about $600.
They also come in a variety of sizes. This one is 5000 watts, so they come in a variety of different sizes. Generators can range in size from this to massive machines capable of lifting an entire building. Generators do what they do because they run on a variety of fuel sources, such as gasoline, diesel, or even natural gas.
You’re going to have to determine what size generators you’re looking for. In critical environments, like hospitals, generators are pretty much mandatory. You don’t want the power to go out and someone to be without a breathing machine. So generators are mandatory in those types of environments. In other environments, it may not be so critical. Maybe you can get away with an “up” if it’s okay to shut your machines off. Perhaps some of your data, as well as some of that redundant data store, is in the cloud.
The other thing here is dual supplies, or dual power supplies, built into systems. Now rack servers in particular are known for this, in which case they will have dual power supplies somewhere in here. Here we go. So I have a dual-redundant power supply unit. You could see that this unit actually doesn’t fit into a tower. This is one of your rack devices.
So this would go into a One-U rackserver, and basically it slips into the server, and you notice it has two power supplies. Power supplies. If you work in the computer industry, you know that power supplies are usually the first thing to fail on a computer, even before the hard drives. In my experience, unless it’s a C-gate hard drive, remember when I was teaching a plus? I have a real problem with T gate. I don’t endorse c gate. But going back to this, power supplies are one of the most famous things to die on a computer.
And when that dies, the machine basically shuts off because there’s no power to it. So this is good to have on servers. That way, they don’t just die like that. One dies; the other one is still working. The server can still stay online. The other term here we’re going to talk about is going to be a managed power distribution unit. First of all, what exactly is a power distribution unit? Power distribution units are basically outlets that manage power. Here’s a power distribution unit. These are going to be rack mounts. They go into racks, and this is where the servers will plug in. Here’s an example of one inside a rack cage, and they just mount in there. Now what happens is that you can see how big they get. Sometimes they go into the side of the rack, and sometimes they go to the top of the rack. It depends on how much power you’re looking for and how many outfits you need. These things are now available in manageable variants.
That is, you can connect to them via software configuration and determine how much power you require, monitor them and keep an eye on them in case they die or have issues, and how much voltage is being used on the system. So you do have a variety of these particular types of units. Okay, so in this video, we talked about quite a lot of different redundancy issues. Remember to always keep your data geographically dispersed across large regions. Don’t keep them all in one region. Load balances are used to distribute loads across many different servers. Nick The interface card team is used to combine Nic cards. Use the UPS to get short-term power. Use a generator to get long-term power. When your power has died, your main power supply is very common on rack servers, especially emission-critical rack servers. These could not be obtained in tower service, but they are not uncommon. And then, of course, make sure you can get the managed PDUs.
2. Replication and storages
In this video, we’re going to talk about the replication of storage-based devices that you should be familiar with. We’ll talk about what our network-attached storage looks like here. We’ll talk about what our storage area networks are, and then we’ll talk about VMs and snapshots.
Okay, so let’s get into this. The first topic we’ll cover is our Nas network-attached storage. So on my desk here, I have our Nas that we use, and this is what is known as the Synology. That’s the maker of this particular Nas. Now you notice it looks like a computer, and it really is a little computer. It’s basically a computer that runs Linux. It has a customized interface, but it’s predominantly made to store hard drives. This thing has 38 terabytes, or, I think, four eight-terabyte hard drives, that we use to store a lot of data, such as these video files that we’re making. So this is a fairly large device in general.
Now, basically, it stores a lot of drives in it, and you could buy this. The price of these devices ranges from slightly more than $500 to $1,000, not including the cost of the drive. and let’s see here. So it’s fairly simple to remove the drives. Here is how easy that was: So this is an eight-terabyte Seagate Barracuda drive that it has in here. Put that back in here. Okay, so this is network-attached storage, right? So it has a NIC card at the back of it. It’s hard to see that there. You guys could see that, though. It has two Nic cards that we could combine to do something. It also has an Eset connector as well as a USB connector.
Okay, so this is network-attached storage. What this thing does is that it basically connects to your network, and you can map drives to it. It’s basically connected to the network. You can map drives, and then it maps it like a big file server on your network, except it’s a dedicated machine made to store data. As a result, many organisations and small businesses will now carry one of these devices. Now big businesses are going to carry something even more complex, something called a storage area network. So this year is not only going to give it—you know what I forgot to mention? These things are highly redundant. This thing runs on a raid. Five. You can run them on raid tens, but you can also run them on raid one. So it’ll already be covered. So don’t forget that these things are highly redundant. The other one we’re talking about is something called a storage area network.
So Storage Area Networks are used in large organisations to store data because they are not only much faster, but they can connect to multiple servers. To give you guys a quick overview of what this looks like, So basically, a storage area network is a big box of hard drives. So it’s a box of many, many hard drives like you see here, okay? But it stores a lot of hard drives. On these big boxes, you can easily get up to 100 terabytes or more. Then what it is is that they have servers, all right? So let’s say server one, server two, server three, and server four, and then they connect them into switches. So that NASA, the San, and the storage area network can all connect to what is known as a “fiber channel switch,” this is now completed. And these are fibre cables that connect to the specific service. Now what it is is that this disc is basically like a drive to the server, but it’s a very, very fast drive. Wow.
This is only a gigabit. Ethernet switches can sometimes run ten gigabits sometimes.This interface is so fast, it’s actually faster than the internal hard drive. Why do we use this? Well, predominantly, the bot organisation puts on this. Databases are the real source of frustration. Databases are very discontinued because every time somebody queries a database or tries to extract data from one, it could take a lot of space. That system might require a lot of CPU power to run. So storage area networks are primarily used in larger businesses. The other thing I wanted to mention—something I briefly mentioned when we did it, and I’m not sure if this video comes after or after—is backup technology. Where I mentioned mostly VMs, we use snapshots. And I said you could do them in VirtualBox, and we’re going to COVID that now.
So, snapshots, what exactly are they? So, what I’m going to do here is demonstrate how to take a snapshot. Notice how this one says snapshot? We’ll boot up one of our Windows 7 computers. Hopefully, they’ll remember that we put viruses on these things and taught them other lessons. So what a snapshot is on a virtual machine is that it’s going to allow us to save a predefined state on that machine. So I’m going to log in here. So basically, it saves the machine. It basically goes “boom,” takes a picture of it at that point in time, and then you could either revert back to that particular snapshot or restart and restore from that particular snapshot. So I’m going to show you guys how to take it, and then I’ll show you guys where you can go and restore it because it does take time to take the snapshot.
So you have your machine running, and you do this because if you’re going to make changes to your virtual machine, if you’re going to be doing updates to it, and you’re not sure if it may crash the machine, take a snapshot before doing it. It’s really easy. We’re just going to go to the machine while it’s running. We’ll say, “Take a snapshot.” and maybe we can say why. Maybe because we’re going to be installing new AV and we’re going to click, okay? And now it goes in and takes a picture of it. Hopefully, it didn’t give me any errors. All right? So it goes in, it takes a snapshot, and I think it’s done already, okay? And then you basically shut it off. Turn it off there. And when you want to view the snapshots, click on this little icon. You say snapshot, and there it is. And then, if you want to start the machine from the snapshot, you click on it.
You would say “start,” and it would start it back from the snapshot. Now, why would you want to do this? Well, because maybe you installed that new antivirus. Perhaps you installed the new system and want to reuse it from there. I’m just going to shut this off because you do have a lot of options here to do the snapshot. So you can double-click it. You can go in there, rename the snapshot, or restore the snapshot. So you can restore it back to that particular one.
Let’s take a snapshot, and then you can restart it if you want. You can clone them out all you like. Why are you doing this? Basically, when you take a snapshot, you’re taking an image, or a portion of an image, of the machine in its current state.
That means no new updates, no new hardware, none.Any other software changes that you make, you can always revert back to them. Once again, if you’re making config changes, the best option is to take a snapshot. Okay? In this lesson, guys, we learned about network-attached storage. We learned about storage area networks, which generally affect big businesses and a virtual machine. You generally take snapshots of them. People also cloned them. By the way, let me show you that before I leave.
Some people also cloned these virtual machines, which I think is a good option. You see, I have a bunch of clones here. You would just right-click on the clone. On a machine, you would say “clone.” And what this does is that, you know, I don’t mind getting another clone here. What this does is let me generate some new Mac addresses.
What this does is that it basically copies everything in the machine: everything in the machine’s file settings, everything. This can take a few minutes. My machine is pretty quick. It shouldn’t take that long, and it basically creates a whole new VM for you. So if you ever install one VM, like you guys see me having a whole bunch of VMs, I generally just install one. I keep cloning it over and over. It’s another way of doing it. OK, hopefully you guys have some fun in this video. The men enter. If you’re using Virtual Box, go ahead, make a few snapshots, make a few clones, and explore how to use Virtual Box.
3. Backup Types
In this video, I’m going to be talking about backup types and different forms of backup. It’s quite an extensive video, but there’s a lot of important information you’re going to need to know for your exam. So let’s get started. When redoing backups, the first thing you should do is discuss with your partner where you will store this backup. Are you doing on-premises backups where you store the data right here? Or are you going to be doing backups where you’re going to be storing it in the cloud?
We do have some differences within that that we should be discussing. You need coffee, guys. When you do it, you need to drink a lot of coffee. Trust me on that. So going back to this on-premises backup, you backup your data. You keep the data within the organization on the same physical premises where you keep it in the cloud. Now, most organizations are probably going to have both, in which case they’re going to store one backup set on premises.
That way, it will be easier to restore any files that are needed by the workers throughout the day and are accidentally deleted. And then, if they have one in the cloud, they may revert to it some of the time. This is the way we do it. This is where a lot of organizations will do it. So you do have the option of doing that.
Now, what I’d like to spend some time discussing with you in this video and what you’ll need to know for your exam is the difference between full incremental and differential backups and what they are. You have to know what these are for your exam. Now, one of the things I want to get out of the way right away is something called an “archive bit,” because this is a term you hear a lot if you read about these types of backups. You’re going to hear a lot about this term, “archive bit.” Let me just show you what it is.
I know there’s a lot more to this, but I just wanted to show you what the archive section looks like. So, allow me to show you the archived bits. I’m going to make a text file here. We call this one the test text file. If I right-click on this and go to properties and go to advance, do you see this? That’s the archived bit. It’s just the file property. You see, if I uncheck it, I click okay, I click OK.
If I open this and make no changes to it, is that right? Look, no changes. I’m not going to touch it. And if I go back and check it, it will still be off because I unchecked it, right? But now, if you change it, add some text in there, and right-click and go to Properties ahead of time, you’ll notice how it turns back on. So what exactly is the archive bit? Well, the archived bit is a bit in Windows that tells you if the file has been modified.
So depending on the backup you’re doing, it will clear that archive bit, letting the computer know, “Hey, those files have been backed up.” In general, when a full backup is completed, the archive bit is cleared. So let’s get started with these three kinds of backups: full incremental and differential backups. Again, you must know the difference for your exam. So let’s go into this here. Okay, so we talked about on-premises and in the cloud. Let’s talk about a full backup. A full backup is one that you generally do once a week because it takes up a lot of space. Assume you have a computer with a storage capacity of two terabytes.
Two terabytes worth of data If you do a full backup, it’s going to take two terabytes’ worth of disc space to back it up. And this is hard to do every single day because, let’s say you want to store your backups for at least one week, you have to back up every day. So if you do a full back up every day with different drives, you’re going to need 14 terabytes worth of disc space. Full backups are not recommended in big businesses because they store a lot of data.
And there’s no point in doing a full back up all the time, every day, because you’re backing up the same thing, the same Windows operating system, and a whole bunch of files that have probably never been changed. Even though a full backup is all you need to do, it is not recommended that you do it every day. Here’s what you need to know about a full backup: When a full backup turns on, it doesn’t care about the archive bit. It just doesn’t care. It backs up every single thing on the computer, whether the archive bit is on or off. And when it’s done, remember, for your exam, you guys should be making some notes. When it’s done, it’s going to clear the archive bit.
So when the full backup is done, it backs up everything and then clears the archive bit. Remember that for your exam. Now comes a differential incremental backup. So here are what these are: So incremental backups are backups that back up everything that has changed since the last backup, and they clear the archive bit. A differential backup backs up everything that has changed since the last full backup. It does not clear the archive bit.
This is an incremental backup that deletes the archive. Let’s say AB is the archive bit. The differential backup does not clear the AB; does it not clear the archived bit, okay? So remember that for your exams. It’s critical to understand. So let me explain how this is going to work, and I’ll give you guys some scenarios.
So let’s say on Sunday, you do a full backup, right? Assume you have a server with a capacity of two terabytes. So you basically have a two-terabyte server. You do a full backup. It backs up every single file on the server. Now, Monday morning comes in, and somebody changes a Word document. Let’s say you only do an incremental backup. Somebody changes the word “document.” So what happens? Well, you come out, and that night, remember, the full backup backed up everything, and it cleared the archive bit. So when the incremental backup starts, it looks for files that have the archive bit on them that were modified. So let’s say you want to change a Word document.
Then it only backs up that Word document. Let’s say the next thing that happens is that you remember. Then it clears your archive bid when it’s done.
Then, when Tuesday came around, you backed up an Excel file. I’ll make this a black one here. It’s difficult to see that when I’m writing inside. Assume Tuesday arrives, you backup an Excel file, and you make a change in Excel. Tuesday night, it only backed up that Excel file. On Wednesday, well, you change the PowerPoint file. It only backs up that PowerPoint file. So it’s like whatever you’re changing that day, it’s backing up only, and then it’s clearing the archive bit.
Now, the thing is, though, if you need to restore the data—let’s say the server crashes—you would have to do a full restore. Then you’d have to do each individual incremental backup.
So you can see, incremental backups are quick because the only backup is really what’s been changed that day. And when they’re done, they clear the archive. But that’s how they know when they come back to follow the only backup. What has changed since that day is the differential? You notice how this is growing. So it starts out the same way with a full backup, all right? So you do a full backup. Differential begins similarly to incremental.
So on Monday, you change the Word document. Again. Just a word. Document. But here’s the catch: it doesn’t remove the archive bit. Then comes Tuesday. Well, on Tuesday, it didn’t clear the archive. That’s the night before Tuesday; you changed that Excel file, but it’s going to back up that Word document again. Why? because it never removed the archive And then on Wednesday, the same concept. You change the PowerPoint. It supports that, as well as Excel and Word.
So you notice how the backup is growing and growing. So remember something: incremental backup, when it’s done, clears the archive bit. Differential backup, when it’s done, does not clear the archive bit. These two backups only back up files that have the archive bit on them. When the full backup is done, it will clear the archive bit on all files. You cannot do incremental or differential backups without doing a full backup. You must complete the full and then one of the others. Now I’m going to show you a piece of software here, a free programme that you guys can download. To perform backups, use ES. and es.com is the domain name. And they have some backup software that you guys can download.
I downloaded the free one here. Where is this one? Backup data recovery Okay, I can’t find this here. I’m going to go to Google. That’s how I found it originally. ESAs backup for free. It’s sometimes easier to browse people’s websites with Google than actually go into their backup. Okay, so here’s the one. I download ESUs to do backups. Notice this is a free download. They do have paid options, but I just got this free one in there. So don’t forget to Google ESA backups. and I’ll show you guys here.
So if I want to do a backup and I’ll show you how you could do it, I’m not going to actually do it because it takes time to do a backup. But I’ll walk you guys through the steps. Let’s say you want to do a backup. Just do a file backup. So here we’ll click on a file backup, and let’s say we want to back up the My Documents folder. So I’m going to click on this, and you want to back that up. And when you realize you need to do it, you simply click Schedule. This is where you would select those incremental differential options. So I would go in here. We want to do a weekly backup. Now notice that if you do daily backups, you have that option, but we’ll do a weekly backup. And it basically starts with a full backup.
And then it’ll start doing either incremental or differential every day after that, every week after that. If you want to do that daily backup every single day, then you do a full backup. And then you could select whether you wanted to do an incremental or differential backup. And then it just keeps doing that over and over and over.So these incremental and differential backup options are what you mentioned in the software. Okay, so don’t forget incremental and differential backups. Expect to see an exam question on that.
Let’s keep on going. Okay, so other terms I have here will be taking a snapshot. Now, snapshots generally apply to virtual machines, right? Generally, you take a snapshot of a virtual machine, and you can then revert back to the snapshot. So there are a lot of different systems, and depending on the actual system you’re using, they would have options to do snapshots. Now, for example, if you’re using things like VMware, they would have the option of doing a snapshot. These more home-based ones, you’re not going to see too many of them to do a quick snapshot in here. However, if you use VMware or Microsoft HyperV, you can take snapshots. Now what are you backing up to? Tape drive or disc drive?
Now, not many companies are using tape drives anymore. Tape drives are very popular because of their mass storage from a long time ago. But nowadays, you’re not going to really see a lot of people backup to tape drives, although some old insurance companies may still have it where they back up to tape drives. The other one is a disc drive. Now most of us know and probably will back up to an additional disc drive. So I may back up my C drive to an F drive, another physical hard drive that I have.
These particular systems, these disc drives, are excellent because this space is inexpensive. You can go out and buy one for under $50, maybe even $30. You can get a seven-terabyte drive for a little over $100 nowadays. That’s a lot of storage you can store on disc drives. Now, copying is simply copying the file over. You can just drag a file to it. If you have an external memory stick, you can just drag a file over to it. Cloud backups are very popular today. A lot of times, we’ll do image backups, snapshots, or PHOTOBOT backups directly into the cloud. Famous companies like Carbonite offer cloud-based backups.
Now create an image backup. Now, image backups are basically what you’re going to use on Windows. When you do a Windows backup, it is basically an image backup. It basically images the entire disc for you. Image backups are very popular in today’s world because they just require a click to restore. You just paste the image back onto the machine, and the machine is good to go. Again, you have online and offline backup solutions. Online solutions indicate that the system is operational and that they can be used instead of offline backups. Now it depends on how you store it. So an online backup is easily accessible and will be mounted, as opposed to an offline storage system, which will not be accessible.
You have to reconnect it. One of the things you want to worry about is distance considerations when it comes to backups. Remember something; there are a couple of things here. There is both good and bad. If the distance is too far to backup, latency may take a long time, especially if you’re backing up from, say, a faraway site, like a site from California to New York. It’s a very big geographical distance for the data to travel. This, of course, can slow it down, but it’s good because if you’re backing up different sites to others, if one site goes down, the other one still has the other data.
So there is some thought to believe that there are both pros and cons to this. Okay, guys, in this video, we talked about some pretty important things you’ll see on your exam. Make sure to know full versus incremental versus differential. On and off, knowledge of the archive grows. Don’t forget, we really don’t use tape anymore, but we do backup basically to disc drives, and we backup basically to clouds a lot of the time. Windows backups are based on image-based backups, and we should consider offline storage when backing up to where the actual main locations are.
4. Non-persistence, high availability, diversity
In this video, I’m going to be talking about non-persistence, high availability, restoration, order, and diversity. This is all related to the redundancy and backups of our systems. So let’s get started. So one term that we should be familiar with for the exam is something called “all no persistence.” Now what exactly is that? That’s when you log into systems and make changes, but they don’t save or there’s no way to reverse them.
And there are various approaches to this. So, first and foremost, some things, such as returning to non-state. One way to do this is to use virtual machines. I showed you guys previously how to use what’s called “snapshots.” If you take a snapshot, you can revert back to a known state. Another thing you could do is, in Windows, do a system restore. So, if you recall, we talked about system restore a few years ago. That’s the way to go back to a particular time and date when there was something that was configured or working well for you. Another option is going to be something called the Last Known Good Configuration.
So you use the Last Known Good Configuration when you boot up your computer and realize it won’t boot because Windows is having problems or has crashed. You could try this. I’ll show you guys how to do this. Basically, I’m going to start up my Windows 7 here, and you just have to press F 8 when it’s starting up to get to the screen. Basically, the safe mode You know what? I missed it. It’s a quick reset here. You know what? I pressed the wrong button. Hold on, guys; I can’t do this right. Here we go. F ten. That’s why I was pressing F9, guys. Okay, so you may have noticed I pressed F8 as the computer booted up.
And if I go down, I have this option here that says “Last known good configuration.” And basically, what this is going to do is boot Windows in a state that it remembers. Hey, these boot files were fine and adequate. Now, there was a problem with my Windows before, but if you ever have a problem with Windows, you could try this option and see if it works for you. It may or may not. Another thing that you can do is boot with what’s known as “live boot media,” and there are a lot of Linux distros that do this. Linux live. You can use these with USB, and then there are a bunch of these types of things, including on boot, to create a bootable USB stick out of this. You could try Buntu before installing this one. Try on Boon to Two before you install it, so you guys can try it.
This is basically a live installation, and what this does is that, basically, you pop it in, and it boots up to a full-fledged operating system while it’s running off of a USB stick or your DVD. It basically runs it all in RAM, and you don’t have to install the operating system. And then, when you’re ready, you just take it out, take the memory stick out of your computer, boot it back up, and you’re back in Windows. So this would be known as a live boot medium. Okay, so the other thing here is high availability. So-called mission-critical systems have to be highly available. Systems with high availability and uptime. This can be done with load, balancers, clustering, and redundant power.
That way, if there is any kind of fault in the system, the systems will remain operational. You want to make sure the system is scalable. Scalability and high availability are really important. Here’s what scalability means when it refers to how large the system can become. As the system grows, high availability becomes more critical. Imagine a website where you only had ten users at the beginning, but as the site started to become bigger and bigger and there were more users and it became more scalable, you’re going to have to have good availability because now you’re affecting more users if it goes down. Another term we should be familiar with is “restoration,” which refers to the process of restoring data. What’s the order of restoring information?
Now, generally, when you’re restoring information on services in an organization, you’re going to always restore the most critical things first, right? So let’s say there’s been a crash on a server, and you have to come and restore the data. And you’re not going to restore data that anyone uses. The restoration order is always from the most critical to the least critical to get them up and running as fast as possible. There are a few factors to consider when it comes to system redundancy and availability. First of all, you want to have a diversified set of technologies. Don’t back up all your data. For example, to just use disk, use disc and the cloud. Don’t back it up, just the tape drives. Use disc clouds.
Diversify the technologies we use as well. Always try to use more than one vendor. Never try to just use one vendor because if that vendor has issues or you can’t connect it to vendor systems, restoring your data could be a problem. Your backups should be encrypted. Encrypting your backups is going to be important because that way, if you ever lose them, no one can just restore them. And then, of course, controls And the controls will be a variety of different controls that will protect your systems. Controls with high availability in particular So high-availability controls are what you would do.
Things like redundant power supplies, load balancers, clustering, rate systems, and just a variety of controls When it comes to high availability, you can’t just think of one thing; you can’t put a RAID system in your desktop and say, “Well, that’s good, but what if the power goes out?” Well, so much for your rating system. You can’t think of just one control, then put an UP on it. That’s great. But when the drive is dead, your computer is too. You need the ups and downs and the rating system. So don’t think of just one thing. Multiple factors contribute to the security of our systems. Okay, so we talked about non-persistent. Those are things that, when you log in, it’s probably not going to change anything. We talked about high availability in the restoration order. Always restore the most critical and then use a diversified set of technology vendors, encryption, and different sets of controls.
5. RAID
In this section, we’re going to be looking at something called raid. Now, raid is a hot topic for your exam, and you should be familiar with the terms for the different types. Now, RAID stands for redundant array of independent (or some people may even say, “expenses”).This is, as far as I know, inexpensive; just go with it. So I think for your exam, it’s more likely going to be independent this.
Now, what exactly is it? Raid basically combines your hard drive and, depending on the type, can give you speed and redundancy. And again, depending on the type of raid, we’ll go through the different types in this section. So by combining your hard drives, you can increase the speed of your data. One of the main reasons they do this is for redundancy; if one drive fails, the other will take over. So I’ll give you an example here, and let’s take a look at the overhead cam. So, let’s pretend you have a motherboard and two hard drives. So I have two hard drives here, right?
My hard drive is 202 gigabytes. So what I would do is connect this hard drive to this motherboard here and one of the ports, right? I need to inspect that cable there, and then I’ll get another set of cables, connect that to this hard drive, and then connect it to the motherboard, correct? And then I’d go into the motherboard and configure this for now, followed by the rate. So, when you connect multiple drives to your motherboard, whether it’s 23467, eight, or whatever drives you have, you simply connect them to your motherboard and then configure rate. Now, you could configure the rate on a piece of hardware, and there are controller cards that you could buy that would go into a PCI slot, one of these PCI slots, like a PCI X 16 slot, and give you the rate that way. Or some of these motherboards come with the built-in raid themselves.
The other way to do rate, and the way I’ll show you in the lab of configuring rate, is to look at how to do it on Windows. So you’re going to connect multiple drives to your motherboard, according to Windows Rate. And I’m going to do this virtually, right? I’m going to look at this on a virtual machine. Instead of having the hardware, the controller on the motherboard or an expansion card, do the rate calculation, you would connect multiple drives to the motherboard and have Windows do it for you. So we want to talk about the different types of raids that exist. Now, I should point out that the rate is not specific to a particular type of drive. Rates for Seta drives could be calculated. You have the option of paying rates for an IDE hard drive or paying for the hard drive itself. You could do rates for MD-2 two drive.
You can do raid for PCIe drives, USB drives, Thunderbolt, and FireWire, as long as the software and hardware support the type of raid you’re doing. Okay, so let’s get into the different types of hard drives and raids that we should know for our exam. And I have some pictures that we’re going to be looking at. Okay, so here are some raids we should be familiar with.
So the first one I have is a rate of 0. For a rate zero, what a rate zero basically does is span the data across multiple drives. So what a rate zero does is that it writes data. I don’t need this yet. So basically, it has two hard drives. And how it works is that it will write data here, here. Assume you’re writing a series of data. Let’s say the cycle is 12345, 6 Let’s say it’s writing ten digits, one to ten. So it’s going to be 123-45, 67, 89, and 10.
So it’s basically writing the dataset across both hard drives. Now, the reason why you’re going to want to use a raid one is because a rate zero is going to allow you to go fast. All right? Read zero really increases the speed of the read-and-write cycle because then it doesn’t have to read. and then wait for it to spin. Then I apologize, write, then wait for it to spin, then write again, then wait for it to spin. So Raid Zero gives you a lot of speed. The problem with it is that in a raid zero, when one drive dies, you lose all your data. So, let’s say you have three drives and a zero, and everything is fine until one of the drives fails. In this two-drive combination, one dies. What’s going to happen is that I’m going to lose all my data. I’m going to have to replace that drive, and I’m going to have to restore all the data from backup.
So you lose that because if you lose one drive, a quarter of your data is gone, so you lose your data. Now the other one I want to talk about is a Raid one. A raiding one is known as a mirror. Remember that a raid one is a mirror for your exam. I like to remember a raid mirror. If you say which one is a mirror, a mirror is like a straight thing, right?
You look in the mirror; it’s a straight thing. I’m looking at the side of the mirror. Okay? So, if my phone is a mirror, then this is one. All right? This is a mirror, and it looks like one. It’s difficult to recall a mirror. So a red one is a mirror. Basically, if these two drives were set up as a mirror, if you got that, if you’re right in that data cycle of ten, you write in the numbers one to ten on our drive. So what it would do is write 123-4567, 8910. I would mirror all that over 123-45, 67, 89, and 10. So both drives contain the same information. That’s why it’s literally called a mirror.
So the reason for that is going to be redundancy, not speed redundancy. So in a mirror, if you lose one of your hard drives, and it’s generally done with two hard drives, it’s no big deal; you’re going to lose. The server is not going to go down. Your workstation is going to go down. You wouldn’t even notice a speed decrease because one drive died, but the other drive had the exact same data as the previous drive did. So in a mirror, a mirror is good for redundancy. It’s not going to give you any speed, though. There is no speed increase.
So what if you say, “Well, Android, I’d like to get some speed boost while also getting redundancy because my hard drive is blowing up a lot?” Now, this is not a bad thing to say about a manufacturer, especially Seagate. I’ve had such a bad experience with Seagate. Maybe you had a better experience with Seagate. I have not had any good experience with Seagate.
In fact, even very large Seagate hard drives, such as a five-terabyte hard drive, a six-terabyte hard drive, or whatever it was the other day, fail on me all the time, and I am stupid enough to keep buying them. Why? I don’t know. I haven’t learned a lesson yet. The next former rate we’re discussing is a Raid Five. So, how does the Raid Five rate the cycle? So a raid five is going to start with a minimum of three hard drives. When you’re installing a raid, you want all the drives to be the same size.
Okay? Then you want all the drives to be the same size. And we’re going to talk about this space between 10 and 1 and 5 in a second, as those are the most unique. So I’m going to add an additional hard drive to my RAID array. Again, I’m going to have 1 TB with 2250s. Let’s just say it’s all 3250s in here. Let’s go back to the overhead cab here.
Here we go with my rating array. So let’s say these three are configured in a rate array. Let’s actually put them all in the same place. So if they’re all configured in a Raid array, here’s how this is going to work: Rate 5 gives you something called a parity bit, and a parody bit is going to be added every time it does one right cycle. So one right cycle would be like that again; those are usually ones, twos, and threes.
That would be like a whole cycle, right? So if this was a zero, if it was 123-4567-8910, and if there was an 1112, keep going, right? That would be a normal rate of 0, but on a rate of 5, rate five is basically a rate zero. It just strips the data across the drive, but it adds a parody bit to every cycle. So a parody bit is going to be used to be real. That’s a special bit that’s used in order to recreate the data. So you write the data in this manner: one, two parity, and one cycle. So remember, it ended at two here, right? ended at two here, and then there was a parody here. The following cycle would be three parity four, followed by another one parity five, six, and finally seven and eight parity.
So basically, every time it goes one cycle across, it’s adding a parity bit in every one of the three cycles with the three hard drives. That’s the good news. The good news about raid five is that because it strips data across the drive, it adds a parity bit while stripping the parity. Remember, the first cycle had the parity bit at the end; the second had the parity in the middle; and the third had the parity on the first drive. So what you want to remember with this is that you could lose one of these drives, all right? You could lose one of these drives, and the rate will keep working. Your system will not go down. The data on the lost drive will be rebuilt using the parity bits from the additional drives. And that brings me to disc space. So I rate zero.
There is no disloc space. If you take the two to 50 gigs I had here and combine them in a rate zero, you’ll get 500 gigs on a mirror at a 50% loss. When you combine these two to create 50 gigabytes on a rate one hard drive, you get a 1 250 rated hard drive. At rate 5, you’re basically going to lose one or a third of your drives. So you have, let’s say, 3250 drives. Let’s see, these are the drives I have here—three to 50 drives, right? And when you add these up, it will actually show up as 500 gigs because you’re going to lose basically a third of the space, which is going to be one of these 250 gigs. So far, the most popular reads are one and five. Now, the other is known as a “read six,” all right? A raid. Six.
Now, a rate six, which is not very common, is this trip, and it’s similar to a rate five, but it adds an extra parity bit in each of those cycles. So it would be four hard drives, and it would be configured in such a way that I could show you this here. Let’s go back to the overhead cam, and we’ll take a look at the rate six. Unfortunately, I forgot to bring the other hard drive. So you know what I’m going to do? I’m going to use my phone. Let’s just say my phone is a hard drive. All right? So now the angle here is getting kind of awkward. Let’s move that keyboard over. So the way rate six would work—and this is not a very common thing—just imagine my phone is a hard drive. So it would go—let’s say the numbers one to ten.
So we’ll go one, two with the data. It would bring parity. Then the second one would be to add the numbers one and two. It would then perform three parity parties and four. Then it would do the third cycle, which would be parity 56. So it’s basically adding an additional level of parity bits. And you could lose more discs like this, in case you want to lose two. Just in case data redundancy means so much to you, maybe you’re running a mission-critical server that, if it goes down, can cause a lot of issues. The news cannot function.
People can’t buy your stuff online, which makes you lose a lot of money. OK, the other one that we’re going to take a look at is a rate zero and a 10 plus one, or one plus zero. So a rate of zero plus one Let’s go take a look at the overhead. Zero plus one is basically a zero and a one combined. So what it does is that it takes my four hard drives again. That other hard drive in this video should have been mine. So basically, it moves the drives apart. So this is a raid zero, and it will mirror everything to this. It technically has to be four drives. It will be 4250-gig hard drives. And basically, it’s a zero here. And this is one complete zero. Then it mirrors this entire zero over here.
So you could lose these two drives completely and still work with this rate of zero. Alright? You could lose this drive completely; you could lose this drive completely. And the rate zero on this set, as well as the rates on this set, can still be used. So when you do a raid ten or a raid zero and one, you could take a lot of disc loss and a lot of different variations of disc lost space. Right, let’s do a quick summary of these particular raids. Here. I have some diagrams, but I decided to show them to you by looking at the driver. I think that was better. Quickly, here’s a quick summary with a table. So the rate zero we discussed can handle at least two hard drives. You can push unlimited hard drives into that. There’s no redundancy; you lose your data, you lose the drive, and you lose all your data. I read one with merit, a driver. That one is like a marriage. Marriage is like both sides of a coin. This one is really only limited to two drives, and one of them may lose data protection. There is no recovery time. You know why? There’s no time.
Once the data fails, the other one just moves over, the other mirror drive kicks in, and no one even knows the drive died. The machine would generally make a beep, alert you, and the drive would fail. According to what I’ve read, five is a minimum of three hard drives that can hold unlimited data. You can lose if you have three. You can lose one drive. As you increase this, you may begin to lose drive. Larger storage provides better data protection. A rate of six is basically a five with a double parity bit. You can actually lose two drives compared to the one you would want to rate five. But you can’t fit four hard drives in there. I saw a one, a zero, and a zero one. So the one I’m showing you is 0 and 1. You can also reverse that with a one and a zero. These are for more performance and more data storage.
This is what you’re going to find more about on the Web site. Okay, so these are rating systems, and we’re going to be following up with them. I did a lab where I showed how to set up software, right? Like in Windows, how do you configure it to do a rate zero and a rate one? with virtual machines. Hopefully, you’ll be able to visit that lab just to practice working with rates. And that’s a Windows rate; that’s a software rate. But most people doing rate are probably not going to be doing software rate. You’re probably going to do a hardware raid, or you’re going to have to do it from your motherboard and get a raid controller card. All right, let’s keep going.
