Welcome to Perusal

What is Cloud Computing?

 What is Cloud Computing?
Perusal Tech Pvt Ltd
  • 07 Sep, 2021

Cloud computing is becoming increasingly popular. It's a significant departure from how businesses have traditionally viewed IT resources. The cloud is being used by businesses of all types, sizes, and industries for a wide range of applications. Even if you don't realise it, you're probably already using Cloud Computing. For example, if you use Gmail to send emails, Google Docs to edit papers, or Spotify to listen to music or watch movies online, Cloud Computing is likely making it all possible. Behind the scenes, not all clouds are created equal, and not every Cloud Computing solution is appropriate for every situation. A variety of models, types and services have evolved to assist you to find the correct IT solution.

 

Financial services firms, for example, are relying on the cloud to fuel real-time fraud detection and prevention. It is being used by healthcare firms to produce more individualised therapy for patients, and it is being used by video game companies to give online games to millions of gamers all over the world. Infrastructure as a Service, Platform as a Service, Software as a Service, Private Cloud, Public Cloud, Hybrid Cloud, Serverless Computing, and many other cloud-related concepts exist. The scale of cloud computing is enormous. There are several cloud providers or firms that provide cloud computing services, on the market today. Microsoft, Amazon, Google, IBM, and others are just a few of the major participants.

 

THE STATE OF IT BEFORE VIRTUAL MACHINES:

 

It is the applications that power and run any business or industry - airlines, tourism, hotels, software, hardware, manufacturing, and so on. Without a dependable, performant, and scalable application, it's nearly impossible for a company to endure pressure and competition and thus stay in business. As a result, the applications are what drive and run businesses. There are two key components to establishing a business application: designing and developing the application itself and then hosting the application on a server to make it available for business. So, let's talk about what we used to do before technologies like virtual machines and containers became available.

 

So, in the past, a firm would need to purchase a physical server to host an application, but the question is how big and fast the server should be. It's difficult to predict the application's demand. To be on the safe side, organisations invest in a high-end server, which is costly. In addition to the initial capital investment of the physical server, we also have monthly ongoing running expenses such as energy charges (both for electricity and cooling), as well as the cost of hiring a staff to operate the server. Don't forget about the operating system licence, which we must also pay for. In many circumstances, just a small portion of the server capacity is utilised, resulting in a significant waste of company funds.

 

As a result, we now have a physical server. We also install the operating system and our business application on top of that. We have our programme up and operating at this stage, and users may use it. Let's imagine we're introducing a new product or establishing a new line of business, which necessitates the development of a new application. So we'll need to get a new physical server, a new operating system licence, and install our second business application on top of that. Keep in mind that we have regular running charges in addition to the cost of the second physical server.

 

You might be questioning at this point if we can install this new application on our existing physical server because the first one is only using a small portion of the server's capacity. In most circumstances, we won't be able to. This could be due to the fact that this new application has different requirements than the previous one. It could be that a different version of the framework is required, or that a different operating system is required entirely. Worse, both apps may require access to the same resources or devices, resulting in resource conflicts. Purchasing a new physical server is one option for resolving this issue.

 

So, in the past, an application and a physical server had a one-to-one mapping. Every time we require a new business app, we wind up purchasing a new physical server, even though these apps only consume a small portion of the server's capacity. What a squandering of company funds. You can see how quickly this may become inefficient and costly for businesses. Until virtualization came along, this was the model we used.

 

WHAT IS A VIRTUAL MACHINE AND WHY IT IS USEFUL?

 

We can get a lot more out of physical servers with virtualization. This is a bit of an oversimplification, but the model operated like this:

 

The operating system is put on top of the physical server. It is the physical server's operating system, often known as the host operating system. We also install a piece of software called Hypervisor on top of that. The Hypervisor is a VMware technology that virtualizes the physical server's hardware. We also develop a virtual machine on top of that. On a single physical server, we can create and run many virtual machines. This is fantastic because we can now properly utilise the server hardware. Between a business application and a physical server, there is no longer a one-to-one mapping. We can now safely isolate and execute many applications from a single physical server thanks to virtualization.

 

As a result, we no longer need to purchase a new physical server every time we need a new business application. We can safely and securely run multiple apps on a single physical server using virtualization. Each application appears to be running on its dedicated physical server, complete with its processor, memory, and disc space. They are running on the same physical server but separate virtual machines.

 

A virtual machine can be thought of as a computer within a computer. On the same physical server, multiple virtual machines can run at the same time. You can install any operating system you like on each virtual computer, including Windows, Linux, and others. In most cases, we install the operating system as well as any dependencies that our production software requires. Virtual machines are quite useful. They enable us to execute multiple apps on a single physical server, allowing us to make greater use of server hardware. However, there is a significant amount of overhead associated with each virtual machine that is built.

 

WHAT ARE CONTAINERS AND HOW ARE THEY DIFFERENT FROM VIRTUAL MACHINES (VMs):

 

We now understand what virtual machines are and how they enable several apps to operate on a single physical server. However, there is a lot of overhead with each virtual machine that we generate, both in terms of cost and maintainability. Let's look at an example to help you understand.

 

Assume we have four apps, App 1 through App 4. We'd like to be able to run all four apps on a single physical server (Server 1). We are unable to run apps on the server hardware directly. We'll need a computer operating system, such as Windows or Linux. At the server level, the operating system is the host operating system. On top of it, we install Hypervisor, a piece of software. This Hypervisor is in charge of creating and managing virtual machines. On this same physical server, we want to execute four apps in isolation. As a result, four virtual machines are created.

 

A virtual machine, as previously said, virtualizes the hardware of a physical server. This means that each virtual machine gets a piece of the server's hardware. We established four virtual machines in our situation. As a result, each VM gets 25% of the server's memory, disc space, processor, and other hardware resources. Our business applications App 1 through App 4 can no longer run natively on a virtual machine.

 

It's important to remember that a virtual machine virtualizes the server hardware. In layman's words, this means that a virtual machine is a slice of the underlying server hardware, and thus an application cannot operate directly on the hardware. So, for a virtual machine to run a software application, it must have a complete copy of its operating system.

 

We have four virtual computers in our scenario. As a result, we'll need to install four operating systems, one for each virtual machine, most commonly Windows or Linux. The guest operating system is the operating system that is installed on a virtual computer. So, we install our business application and any dependencies it may require on top of this guest operating system on the VM. Consider the server resources necessary to install and execute four different operating systems.

 

DOWNSIDES: To keep these VMs up and running, a lot of server resources like disc space, memory, and CPU time are wasted. In addition to the squandered server resources, each operating system installation necessitates administrative time. For example, if an update or security patch for the operating system is available, the administrator must download and install it on all impacted VMs. We also have to pay for four operating system licences on top of that.

 

Keep in mind that a virtual machine is essentially a computer within a computer. It comes with its operating system. As a result, for a virtual machine to be operational, its operating system must also be started. It takes a long time to boot up a full operating system. If we want App 1 to be up and running, we must boot up the VM 1 operating system in addition to the host operating system, which is the operating system on the actual server.

 

CONTAINERS: Let's look at what containers are and how they work to tackle these problems:

 

(1) Remember that a virtual machine virtualizes the server hardware, whereas containers virtualize the operating system, therefore it's an abstraction at the OS level. On the same machine, many containers can execute and, more significantly, they can share the host operating system kernel. A container, unlike a virtual machine, does not require an operating system. So you can tell how much disc space, RAM, processing time, and other server resources are saved right away. Each container encapsulates your application's code as well as its dependencies. You construct three containers if you want to run three apps on a single physical server. Although each application believes it is running on its operating system and server hardware, all container applications are sharing the same host operating system and hardware.

 

(2) Virtual machines take admin time to deploy security updates and keep the host operating system and guest operating systems on all VMs up to date. In the case of containers, there is only one host operating system that requires admin time, regardless of how many containers we have on our server.

 

(3) In terms of cost, we must pay for the host and guest operating systems of all VMs in the case of VMs, however in the case of containers, we only have to pay for one host operating system. In comparison to a virtual computer, a container is quite small.

 

Because a container does not need to boot up its operating system, it can usually be brought online in a fraction of the time it takes to boot up a virtual machine and its operating system.

 

BENEFITS OF CLOUD COMPUTING:

 

(1) COST SAVINGS: One of the primary advantages of switching to the cloud is lower IT costs. Before cloud computing, if an organisation developed a business application and wanted to make it available online for users, it had to purchase a physical server and all supporting infrastructure to host the programme. Servers are costly, and firms must incur a significant upfront Cape or capital investment. It doesn't end there, though.

 

We need a server room and that needs to be secured, set up and configure the server, connect to UPS and network, ensure the proper cooling system is in place, install the server, operating system, antivirus software, database server, and any other dependencies your business application may have, such as dot net framework and a web server like IIS, for example. This is just the beginning. It also doesn't end here.

 

We must keep these servers up and operated, i.e., in good working order. It is necessary to download and install any updates or security patches that are available. We are again liable for replacing failed hardware in the event of a hardware failure, power outage, natural disaster, or other crisis. What about backups of databases, disaster recovery, failover systems, and so on? Both for the initial setup and continuing maintenance, we'll require a trained IT team.

 

So the point we're trying to make is that businesses are faced with a significant Opex, or running expenses, which occur every month. Both Capex and Opex are reduced when you go to the cloud. Physical servers, related IT infrastructure, professional IT workers, server rooms, and data centres are no longer required for organisations to spend large sums of money. Rather, we use cloud infrastructure and pay a monthly subscription for the resources and services that we consume.

 

(2) PAY AS YOU GO: The cloud is a pay-as-you-go model, which means you only pay for what you use. It's similar to your utility bills, such as your power or water bill. If your application has a lot of demand and you utilise a lot of cloud resources like storage and computing power, you'll have to pay extra. You pay less if you use them less. To summarise, cloud resources are metered, and you only pay for what you consume.

 

(3) SCALABILITY: Cloud computing offers a great deal of scalability and flexibility. If you're starting a new business or product line and need more cloud resources, you can easily scale up cloud resources like memory, processing power, and storage with a single click. Setting threshold limitations can potentially automate this process.

 

For example, if 90% of the current storage capacity is used up, add additional 100 gigabytes to the storage. You can also do the opposite. You can reduce the resources if you aren't using them. You can specify the threshold limits and automate even while scaling down.

 

We will never run out of resources with the cloud. Scale-up when you need more power and down when you don't. You are no longer responsible for obtaining and installing those costly upgrades. This is handled by your cloud services provider, like Azure or Amazon Web Services.

 

(4) ACCESSIBILITY: Cloud-based applications, services, and data can be accessed from practically any location and at any time, whether at home, at work, on vacation, or while commuting. All you need is a gadget that can connect to the internet.

 

(5) BUSINESS CONTINUITY: The cloud ensures that your business runs smoothly. If your data is stored on an on-premise server, and there is a hardware failure, power outage, or other disasters, your data will be lost. It is exceedingly difficult to run a firm without data. The truth is that without data, there is no business. If you have data in the cloud, on the other hand. Data loss is not caused by hardware failure, power outages, natural disasters, or any other crises. This is due to the fact that data in the cloud is backed up across multiple networked servers.

 

(6) AUTOMATIC UPDATES: The cloud ensures that updates are made automatically. You are responsible for obtaining and installing software updates and security patches if you have your services and apps on-premise, which is not only unpleasant but also time-consuming. If you use the cloud, however, security fixes and upgrades are automatically installed by the service provider off-site. All you need to do is pay a little monthly charge. You won't have to worry about updating and patching your system yourself. Instead, you can use the time to concentrate on what is important to the company.

 

(7) BETTER COLLABORATION: The cloud makes it easier to work together. Teams and team members may access, modify, and share files anytime and from anywhere, thanks to the cloud. There's no need to submit documents or spreadsheets over email any longer. It's all stored in the cloud, and anyone who wants to use it must do so. As a result, teams will be able to better collaborate and work together.

 

(8) SELF-SERVICE: The majority of the cloud provider's resources and services are self-service, which means that most people can easily acquire, configure, and use cloud resources and services with a brief learning curve. If you need a server to host your company application, for example, you don't need to visit an IT specialist to figure out how big and fast the server should be. You may easily obtain a virtual machine from the cloud and host your application with only a few simple clicks. You can scale up if you require more processing power, and scale down if you do not.

 

RISKS OF CLOUD COMPUTING:

 

(1) LOSS OF CLOUD DATA AND SERVICES: One of the most significant hazards is the loss of cloud data and services. Data and services in the cloud are often accessed over the internet. As a result, the speed with which you can access cloud services and data is constrained by the speed with which you connect to the internet. If you lose your internet connection, you lose access to cloud data and services as well. The cloud service provider's servers or network can also fall for a variety of reasons. Again, you will lose access to cloud data and services if this occurs. The good news is that cloud service providers have service level agreements or SLAs. The majority of cloud providers offer 99.9% uptime.

 

​​​​​​(2) DATA SECURITY: When you store your company's, employee's, and customer's data in the cloud, you're putting your faith in the cloud service provider to keep your data safe and secure. Cloud service companies such as Microsoft Azure and Amazon Web Services devote a lot of time and money into implementing and improving cloud security, but storing your data in the cloud rather than on your premises exposes you to security concerns.

 

Furthermore, everyone, even hackers, understands that if they can hack into cloud service providers, they will gain access to massive amounts of data. As a result, there are hundreds of thousands of cyberattacks taking place right now all across the world. So the point we're trying to make is that when you put your data in the cloud, you're relying on the cloud service provider to protect it, and your data is only as safe as your cloud service provider. Best practises such as encryption, two-factor authentication, auditing, reviewing, and rotating access keys and credentials can, however, help to mitigate the security risk.

 

(3) COMPLIANCE AND LEGAL RISKS: Another key risk to consider when transitioning to the cloud is compliance. You don't have to worry about complaints or legal issues if it's your data, such as emails or images. If it's an organisation that deals with financial data, healthcare data, credit card data, or any other regulated data, the organisation is required by law to protect that data.

 

It also needs to know where the data is kept, who has access to it, and what safeguards are in place to protect it. Many local and international rules exist, such as GDPR and HIPAA. By migrating to the cloud, you're putting your trust in the cloud service provider to protect your data and comply with local and international requirements. If the cloud service provider fails to deliver and there is a data breach, your firm may still be held accountable.

 

(4) COST CONCERNS: Cloud computing cuts upfront infrastructure costs and offers greater flexibility due to its pay-as-you-go concept. The overall price you pay is determined by traffic, the number of cloud resources you utilise, the plan you choose, and how you scale resources up and down. The total cost may be higher than you anticipated.

 

Most cloud service providers, such as Amazon Web Services, Microsoft Azure, and Google Cloud Platform, have cost calculators, which you should utilise to get an estimate of what you'll pay. You can also try out other alternatives and plans until you discover something that works best for you. There are a few things you can do to cut down on your cloud fees.

 

Try not to over-provision cloud resources, for example. If you suspect that performance will be an issue, you can always auto-scale up resources depending on pre-configured threshold limitations and don't forget to auto-scale down. You'll end up paying for resources that aren't being used if you don't.

 

Similarly, automate the process of starting and stopping your server instances. This ensures that you are not charged for services you are not utilising. You may also set up notifications to keep track of your cloud spending. Before shifting to the cloud, it's always a good idea to bear these hazards in mind. The good news is that various types of clouds, as well as best practices and tried-and-true methodologies, can help to mitigate some of these dangers.

 

STAY TUNED TO LEARN MORE ABOUT CLOUD COMPUTING.