Public Cloud Beta

A public cloud is a cloud infrastructure organized by an ISP and presented to multiple users over the Internet as a service. In a public cloud, the provider's shared computing resources are shared among different users.

The benefits of using public clouds:

• Scalability.

User resources consumed in the public cloud can easily and quickly scale without having to incur capital costs, the user only pays for the resources actually consumed. During peak workloads, resource consumption can be increased, and when resources are not needed, they can easily be discarded, reducing costs.

• Reduced costs for short-term projects and low entry threshold

With small amounts of service consumption and short periods of time using cloud infrastructure, the public cloud can reduce costs. With a public cloud, you do not need to purchase and organize physical infrastructure, buy cloud infrastructure software, and hire and manage support staff, making it a good option for startups, where project timelines are difficult to predict and high capital expenditures on equipment are unacceptable.

The cloud provider is responsible for protecting the hardware and software of the cloud infrastructure, separating access to resources and authenticating users, while the customer is responsible for the security of their own resources: virtual machines, applications and data on virtual machines.

Clouds are often, more secure than small private data centers because cloud providers have the resources to hire enough highly trained staff, automate operations, audit and upgrade security systems. Cloud infrastructure providers have the tools to isolate workloads, encrypt data and detect potential threats.

Typically, large data centers that host the physical infrastructure of a private cloud have a number of tools to improve system resiliency:

• Power redundancy.

Achieved by connecting infrastructure elements to two power lines that are independent of each other. Load switching between the main and backup power supply is done automatically without any power failures. When the main power supply is unavailable, back-up power supplies, such as battery systems or generators, are used.

• Communication redundancy

The data center is connected to multiple Internet service providers at the same time to ensure uninterrupted communication.

• Redundancy of resources

If a server goes down, the virtual machines immediately start up on another server. As a result, the cloud continues to run without downtime, and services and resources remain available to users.

• Data Replication

Replication is a mechanism for synchronizing the content of multiple copies of an object. In replication, changes made in one copy of an object can be propagated to other copies, which prevents data loss in the event of a failure of the physical media on which one of the object copies resides.

• Migration during maintenance

In the case of equipment maintenance, an automatic seamless migration to other equipment is used. This process allows equipment maintenance to be performed in accordance with the manufacturer's regulations.

• Fault tolerance at the application level

Utilizes services for dynamic routing, traffic balancing, API command broadcasting to infrastructure, etc.

• Fault tolerance

Large cloud providers have multiple data centers in different regions, allowing them to redistribute load among other data centers in the event of a failure in one data center.