Universal Server in 2025

A Universal Server is a device designed to address a range of tasks that can be fundamentally different in terms of software architecture and system requirements. The configuration of a universal server is tailored for the maximum performance of all components to ensure efficiency in various operations.

Tasks of the Universal Server

Sometimes, purchasing server hardware for each specific task can be costly. It’s simpler to consolidate everything within one device. Even for home use, there are certain tasks designated for a device capable of handling them. So, what tasks can such a device tackle?

1. Terminal sessions: Allocating remote desktops, working with applications in terminal mode.
2. Video surveillance: IP cameras and their management.
3. Remote access to the local network.
4. Virtualization: Though attempting this within such a configuration is not advisable.
5. Hosting: Allocating resources for websites or web services is entirely feasible.
6. File server.

However, why am I listing all this? Let’s put it more simply: it can address any tasks that servers can handle but comprehensively.

Consider a common scenario: working with graphics in terminal mode, needing powerful graphics cards, concurrently running 1C in terminal mode, deploying a small cloud with file-level access. And this whole mix is accessible via the internet, with an additional security system in place.

A shining example of a universal server. All in one.

But, of course, when solving such tasks, there are no easy solutions. Why? Because there are certain pitfalls associated with users. After all, if all resources are utilized simultaneously, the equipment will have to cope with a substantial workload.

Technical Specifications of the Universal Server

So, let’s break it down by tasks:

1. For 1C in terminal access, a processor with a high clock frequency is required. The number of cores beyond 4 doesn’t provide significant advantages. The RAM size should match the size of the 1C database. Depending on the number of users, you may need to choose disk drives. HDD is good for 5-10 users; if more, it’s better to opt for SSD, ideally with an NVMe interface. It’s not recommended to use universal servers with a large number of users – the more people, the more resources need to be allocated for the 1C server. Only if there’s a universal cluster, but that’s already impractical. It’s simpler in a cluster to allocate a separate device for a specific task.
2. Working with graphics requires the presence of graphics cards. Depending on the tasks, for example, drawing and editing photos in a graphics editor don’t require serious resources, but applications like Photoshop have substantial requirements. However, 3D modeling, working with VR, and rendering will consume the resources of the graphics card, and a portion of the processor’s power will be needed to support the graphics card’s operation. Sufficient RAM and disk space are needed for program deployment. For instant recording of rendering results, it’s recommended to use solid-state drives.
3. Cloud with small applications like text editors requires minimal technical resources, but a file server will require a portion of the disk space.

Seemingly nothing special, but implementing it is quite challenging. So, we’ll need a powerful processor, preferably 2. Why? Well, honestly, up to 10 users will be fine with a single central processor, but for 20 or 30 users, it’s better to allocate a separate one.

So, for 20-30 users, using Intel Xeon Silver 4416 and above will be sufficient. And yes, install two of them. Of course, if scalability is planned, it’s better to use processors from the Gold series, as Silver doesn’t support using more than two processors.

One CPU will be for 1C, the second for graphics work and other tasks. Insert a stack of graphics cards, not necessarily professional ones, but those that can handle the tasks well. Servicing all this setup will require a power supply of 1200W or even 1500W, depending on the number of processors and graphics cards.

And yes, a key point is software and the database on SSD, while other things, like file storage and miscellaneous, are better off on separate HDDs. The number of drives will depend on RAID.

The more users and tasks, the more components you’ll need to install. For example, if you add terminal access to any application and video surveillance to this, you might already need a four-processor system.

In reality, just adding users is sufficient.

Efficiency

The key point is efficiency. The more users for each task, the more resources are required. A universal server can be assembled only within the average load range. The majority of technical resources are directed towards the execution of the primary task, with fewer resources allocated for ancillary tasks.

Certainly, it’s possible for several tasks with approximately equal loads or in cases where they are performed with different frequencies. For example, during the day, accountants work with 1C, and in the evening, designers log in to create layouts. In general, the essence is clear.

Now, let’s assume that we need a universal server for 30 employees engaged in developing 3D models, 200 users of 1C, and another similar number of programmers. Realistically within the confines of a single server?

I doubt it. Even for 200 programmers, a cluster might be required. Code processing, analysis, compilation – any programmer understands what recompiling a few thousand lines of code entails. What about several hundred thousand lines? Programmers probably shudder at the thought of such processes.

A programmer might rewrite and initiate the compilation process dozens of times a day until the code error is corrected or an acceptable result is achieved. In general, 200 such programmers can easily overwhelm even the most powerful server with 4 CPUs.

200 users of 1C will also consume a lot of resources. In this case, it’s almost certain that a separate server for the database will be needed.

In essence, a universal server is:

• Multiple tasks with light loads.
• Sufficient resources to meet requirements comprehensively or individually.
• Everything implemented within a single device.

Actually, nothing complicated. Hence, I say that the most vivid example of a universal server is a home server. Many small tasks, but all consolidated in one “box.”

Security Issues

Every software, every operating system has its unique vulnerabilities. Let’s assume a vulnerability is found in the software, exploited to target the universal server. Done, the device is compromised. The functionality of the entire department is undermined. Well, or a large amount of data is stolen. In general, there are numerous variations.

Certainly, complex schemes with process isolation from each other can be implemented, but it’s simpler to use different devices for different tasks and departments.

So, if security is a crucial concern, it’s better to refrain from such IT infrastructure construction and prefer individual devices.

In Conclusion

I hope I’ve managed to explain what a universal server is in an understandable way. It’s more of a principle of structuring than hardware, but do you need such a construction? Decide for yourself. It can be very effective or not suitable at all. The main thing is that after reading this, you find it easier to make a decision.