We use cookies
We use cookies to optimize our website. By continuing to browse the site, you agree to our use of cookies.
New design
Control panel
  • Русский
  • Українська
  • English
  • UAH
  • USD
  • RUB
  • EUR
  • 0-800-307-307 Hotline
  • +38 (044) 392-74-33 Kiev
  • +38 (057) 728-39-00 Kharkiv
  • +38 (056) 794-38-31 Dnipro
  • +38 (032) 229-58-93 Lviv
  • +38 (048) 738-57-70 Odessa
  • +38(093) 170-15-42  Life
  • +38 (067) 400-88-44 Kievstar
  • +1(888)393-24-51  USA, Toll free
  • +44(131)507-01-14  Great Britain
  • +7 (499) 348-28-61 Moscow

How to choose the right dedicated server?

Dedicated server or Dedicated server – this is a type of hosting, in which the client is provided with a separate physical machine entirely. When placing a web resource on a dedicated server, one client has all the power of the machine at full disposal, there is no need to share the server resources with neighbors, as is the case on virtual VPS / VDS servers.

Typically, a dedicated server is used to run applications that have high resource requirements or, for some reason, cannot coexist on the same server with other projects.

Having received a dedicated server for rent, you can independently select components, choose a processor, RAM, disk space, and connect additional network equipment or other specific hardware. You can create a configuration that will best meet the technical requirements of your project and will provide the best performance regardless of the complexity of the tasks.

What should you consider when choosing a dedicated server?

By choosing dedicated server, ask yourself the questions: "What will this server be used for and what technical characteristics are needed for this?", "What are the requirements for this project to the equipment?" Perhaps it will be an online store with high traffic, or a platform for developing web applications, or you plan to store a large amount of data on the server – all this must be taken into account when choosing hardware, since in some cases you will need a large amount of RAM, and in others a high-frequency processor. Let's figure it out.

CPU

Processors have two main characteristics, the number of cores and the frequency.

In multi-core processors, as a rule, the frequency is reduced, this allows you to reduce the power consumption of the processor without losing performance. These processors are used in processes that support multithreading and parallel computation, such as virtualization or video encoding. An example of such multi-core Intel Xeon Gold and Intel Xeon Silver processors.

For some tasks, on the contrary, a high core frequency is important, while their number is less important. High-frequency processors are suitable for servers on which licensed software is used (for example, Windows OS) with a core rating, or if you have a 1C database installed on your server. An example of such a processor would be the Intel Xeon E-series.

RAM

The RAM must meet such important requirements as high reliability, maximum performance, low power consumption.

In a server system, RAM is responsible for the speed of data processing. It temporarily stores the output, input and intermediate data of the operating system and programs. Therefore, when choosing memory for the server, you need to remember that the more calls to the server, the more memory is needed to save this data, and if there are not so many calls to the server and there is no need for high speeds, then you should not put a lot of memory, you can increase capacity as the needs of your project grow.

Bandwidth of network interfaces

Interface bandwidth is the maximum rate at which data can be transmitted over a given interface.

ISPs typically offer 1G (1Gbps) and 10G (10Gbps) Ethernet ports.

1G ports can be used to access the Internet, their bandwidth will be enough for this, and as a rule, all servers have 2 such ports built in.

10G ports are more high-speed, they are used for organizing a local network and they need to be installed additionally.

Disk drives

For disk drives, the following parameters are important: access time and capacity.

SATA drives have read speeds significantly slower than SAS and SSDs, but they can store large amounts of information (up to tens of TB). Such disks are suitable for large volumes of static content, backups, video encoding, for game servers, as well as for large, but not loaded file servers – for all those tasks that do not require frequent writing and reading information.

SAS disks have high performance. Their rotational speed is twice that of SATA, so they should be chosen for tasks that require high speed and multi-threaded access. Also, SAS drives (unlike SSDs) can provide reliable and multiple overwrites of data, which makes them very reliable, but unlike SATA, they have a small volume. SAS are well suited for hosting, database management systems, distributed systems, high load WEB servers, terminal and 1C servers.

SSD drives have high mechanical durability, reduced power consumption and high operating speed, but at the same time they are limited in the amount of information that can be rewritten to the disc. Accordingly, if your system will overwrite more than 20 GB of data per day, then under such a load, the SSD will need to be replaced after some time.

The use of SSD drives is suitable for any high-load projects. These can be loaded sites, sites running on modern CMS, which require simultaneous access to several files on the disk when generating a page. They are also suitable for storing databases and a boot partition.

In order to ensure the safety of data, it is worth choosing server configurations with SSDs in RAID arrays. This will greatly increase the fault tolerance and speed of data writing / reading. In the event of a failure of one or more disks, depending on the RAID level, you will still be able to access the data.

RAID arrays combine multiple disk devices into a single unit. Choose a server with RAID1 or RAID10 arrays.

RAID 1 is the simplest fault tolerance configuration where data is copied from one disk to another at the same time, creating a mirrored copy. The advantage of this configuration is that if one disk fails, the other will continue to work. The downside is that performance is significantly reduced, since the volume of disk devices is halved.

RAID 0 is used to improve server performance by writing data across multiple disks to improve I / O. This configuration requires a minimum of two disks. The downside is low fault tolerance. In this case, failure of one disk may result in data loss or damage, as this will affect the operation of the entire array.

A RAID 10 array is a combination of RAID 1 + 0. This configuration provides high performance and is ideal for heavily used database servers or any server that does a lot of writes. But you need to understand that to implement such a configuration, you will need twice as many disks.

Any of the above configurations can be implemented through software or hardware, but it should be borne in mind that implementation in the form of software entails the loss of many performance benefits.

 

RAID 10 array

 

The importance of backups

No matter how hard you try to ensure a high degree of safety of your data, the risks of losing information, albeit minimal, will always remain. Even if you choose a configuration with several RAID arrays for your server, the RAID controller itself may fail, and many other factors can affect the safety of your data, therefore we strongly recommend using backup system and store backups on a separate storage.

Choosing a data center. Reliability comes first!

If you have decided on the components and configuration for your server, all that remains is to choose a reliable data center. When choosing between hosting companies, you need to study the proposals: what is included in the services, what technical characteristics and indicators are declared on the site, whether their data center meets all the necessary security requirements, is there any means of protection against hacker attacks and unauthorized access to data even in the event of a failure server, whether backup is provided, and it will not be superfluous to get acquainted with the reviews of clients who have been using these services for a long time.

Here are the main indicators of data center reliability and their best characteristics:

  • fault tolerance – 2N;

  • distributed streams – 2; 

  • uptime – 99,99%;

  • idle (h / year) – 0,4;

  • scheduled work without stopping the data center – Yes.

There are 4 levels of reliability (Tier) and each of these levels corresponds to different indicators. The data that we have given above corresponds to the highest level of Tier 4. Tier 3 is still acceptable, and Tier 2 and Tier 1 are already a bad choice, since a decrease in the described indicators indicates a lower level of data center reliability.


Join Hosting Ukraine and you will receive:

  • Quality and reliable hosting with a good price-quality ratio: SSD hosting, VPS on SSD, Cloud (cloud) hosting, cloud VPS. Dedicated server or dedicated server.

  • Domain registration in 5 minutes.

  • The ability to install a suitable CMS in one click. Joomla, Wordpress other popular CMS with PHP, Perl, MySQL support.

  • Convenient management through the admin panel - everything in one place.

  • Intelligent protection system.

24/7 technical supportto help resolve any issues you may have.


Only registered users can leave comments

Subscribe to the newsletter

We will send announcements of new articles and useful tips once a week

Follow us on social networks