Skip to content

Dear customers! Please send orders and inquiries to [email protected] or to Telegram @servix_uz instead of calling by phone: due to the high workload, the operator cannot answer all incoming calls.

Servix

How to choose a server: 1C, virtualization, file server, database

Servix technical team · Updated

In brief. A server is chosen for the workload. 1C needs a high-frequency processor, often a single socket, and only SSD or NVMe: DL320 Gen11, DL360 Gen11 or Dell R660. A virtualization host requires two multi-core sockets, memory with at least 30% headroom and two 10GbE ports: DL380 Gen11 or R760. A file server is built on high-capacity HDDs with RAID 6, while a database requires NVMe, maximum memory and a second machine with replication. First, pin down the number of users, the data volume for 3–⁠5 years and the acceptable downtime.

A mistake in choosing a server is expensive: an insufficient configuration hits its ceiling within six months, while an excessive one freezes the budget in unused resources. Here is what to start from when selecting a server for the four most common scenarios.

Start not with hardware but with four figures

Before discussing models, gather the input data:

  1. Number of users and load profile. 30 accountants in 1C and 300 employees in a CRM call for fundamentally different machines.
  2. Data volume today and the forecast for 3–⁠5 years. Plan for the server's lifetime, not for the current moment.
  3. Acceptable downtime. An hour, a day or zero — this determines whether a second server and a cluster are needed.
  4. Placement conditions. A rack in a data center, an office server room or a utility room — this is the choice between rack and tower.

Server for 1C

1C is sensitive to clock speed and latency, not to the number of cores. A processor with 32 slow cores will lose to one with 16 fast cores.

  • CPU: high-frequency models; one socket is often enough.
  • Memory: with headroom — the DBMS database must fit in the cache. Saving on memory here makes no sense.
  • Drives: SSD only, preferably NVMe. HDDs for a 1C database mean guaranteed user complaints.
  • Platform: HPE ProLiant DL320 Gen11, DL360 Gen11 or Dell PowerEdge R660 — for a small or medium-sized company. Tower ML30 Gen10 / ML110 — if there is no rack at all.

Virtualization host

Here it is the other way around: what matters is cores and memory capacity. Memory almost always runs out before the processor does.

  • CPU: two sockets with a high core count. Take into account the hypervisor licensing model — it is often calculated per core and affects the total cost more than the hardware does.
  • Memory: the key resource. Populate all channels and take at least 30% headroom.
  • Storage: either an external storage system or local NVMe for hyperconvergence.
  • Network: at least two 10GbE ports, preferably 25GbE. Separate management, production and storage traffic.
  • N+1 rule: the cluster must survive the failure of one node without degradation. Calculate cluster resources with this reserve, otherwise a node failure will stop some of the virtual machines.
  • Platform: HPE ProLiant DL380 Gen11 / DL360 Gen11, Dell PowerEdge R760 / R660.

File server and backup

The task is bound by capacity and reliability, not by the processor.

  • CPU: entry-level models; one socket is enough.
  • Drives: many bays, large SATA/SAS drives. For backups, HDDs remain economically justified.
  • RAID: RAID 6 or RAID 60 for large drives. RAID 5 on high-capacity drives is dangerous: during an array rebuild, the failure of a second drive leads to complete data loss.
  • Platform: DL380 Gen11 in a configuration with 12 LFF bays, Dell PowerEdge R760xd2.
  • Important: a backup on the same server as production is not a backup. You need a separate site or at least a separate device.

Database server

  • Drives: NVMe and only NVMe. The disk subsystem is the main bottleneck of a DBMS.
  • Memory: as much as the budget allows. The active part of the database must live in RAM.
  • CPU: depends on DBMS licensing — for a number of commercial systems, the license is calculated per core, and “extra” cores cost more than the server itself.
  • Fault tolerance: two machines and replication. A single server for a critical database means accepting the risk of downtime for the duration of repairs.

Three things people forget

Power and cooling. A dual-socket server under load consumes noticeably more than the datasheet suggests. Check the UPS and room cooling headroom before the purchase, not after.

Manufacturer support. A 3-year warranty with next-business-day response and a “bring it to the service center” warranty are different levels of service availability.

Expandability. Free memory slots and drive bays are the cheapest insurance against an error in the load forecast.

Describe your task at [email protected] or in Telegram — we will select a configuration, price out the new and refurbished options, and help with installation and setup.

Related selections and tools: Server for 1C, Server for virtualization, File server and backup server, Dell PowerEdge.

Related reading

Rack or tower server: which to choose

Rack or tower server: a 19″ cabinet or an office, noise from 22 to 49 dBA, 1U and 2U, which HPE and Dell towers fit into a rack and what is cheaper with the same components.