Short and sweet: how to choose a RAID/HBA controller for the server
RAID controllers have not left and are not going to, on the contrary, hardware controllers for NVME and PCI Express drives (including M2 format drives) are entering the market. The abundance of drives and software features of RAID/HBA make you approach this choice as responsibly as possible, otherwise you may lose data.
RAID controllers have not left and are not going to, on the contrary, hardware controllers for NVME and PCI Express drives (including M2 format drives) are entering the ...
Server fans: what they are, how they work
Server fans play a key role in maintaining the temperature at an optimal level. They operate at high speeds to ensure sufficient airflow within the server system, thereby cooling it. Most server fans are designed to operate at different speeds, which significantly enhances their efficiency. This means that when the server system is under load, the server fans operate at higher speeds to increase cooling, and when the load is light, the speed decreases.
The main function of server fans is to ...
Lenovo ThinkAgile VX3320 review - a hyperconverged solution for simplified enterprise-wide vSAN deployment
Hyperconverged systems are rapidly gaining popularity and are already becoming the standard for building virtualization platforms in the corporate environment. Lenovo has released a special series of servers for installing and running VMware vSAN with tools for simplified administration and deployment at the customer's facility.
The ideology of hyperconvergent solutions using VMware technologies has its advantages: this is a relatively low cost at the start, and the ability to scale almost ...
Review of Lenovo ThinkEdge SE50 and SE30-industrial and peripheral PCs based on Core i7 with passive cooling
These are universal computers that can simultaneously be used in production, at peripheral facilities, and are equally well suited for PC-controlled equipment, for building control rooms, laboratories, and engineering workplaces. No moving parts concept, 4G/LTE support with eSIM, DIN rail mount and RS232 ports.
The development of robotics and automation of workplaces today has led to the fact that the OT industry (short for operational technologies), which includes a system for receiving ...
Review of the 4-CPU Lenovo SR860 V2 server based on Intel Xeon Cooper Lake
... MHz
four 28-core Intel Xeon Platinum 8380H a total ... and after all, each Intel Xeon 8380H consumes up to 250 ... processors, which include the Intel Xeon 8380H installed here. The ... server in question is, of course, 4 high-performance Intel Xeon ... we need to understand the Xeon Platinum 8380H frequency matrix.
... advocated the use of the TensorFlow and PyTorch frameworks on the ... machines.
Note that Intel Xeon processors with an H index ... all the documentation for the server, Lenovo calls the memory the word ...
Article about vertical wall-mounting of servers
An interesting solution for home laboratories, small test rooms or maintenance-free points where 1-2-3 pieces of equipment are placed, and every square centimeter is worth its weight in gold
I often came up with the idea: why not place servers vertically, because if you do not have a special server rack, but have a 1U-4U machine, and even a full length of 655 mm, then this bandura will take up the entire room, the entire table or rack. No, I'm not saying that server cabinets are outdated and ...
What factors should be used to choose the server processor?
Why when choosing a processor, do not pay attention to the price per core or the price per Watt, but rather use the cost of a Megahertz with HT disabled and using Turbo Boost?
Here's how it used to be good: for servers, there was a single Pentium Pro processor, which you didn't need to choose. With the advent of multi-core processors, it became clear that the more computing blocks, the better, but only until marketers stepped in and pided one CPU into ten sockets with ten cache options and ...
How water displaces air in Lenovo supercomputers
A couple of years ago, the maximum air cooling limit was 600W per unit. Today, it is theoretically possible to divert up to 1.2 kW per unit using air. However, in the future, you will need to allocate 2 KW, and you can't do without liquid cooling.
Today, there is a steady trend to extract the maximum power from each computing unit of the data center. Users want to see more and more local disk storage inside data center nodes, and circulatory data transfer systems that combine cluster ...
Review of ASRock Rack E3C242D4U2-2T motherboard for budget office server
Practically, when you need a solution for a host running a security gateway, a 1C server and a Kubernetes node, you can even give up ECC memory in favor of Core i5, Core i3 and even Pentium Gold. In general, where a high frequency is needed in single-threaded loads, it is often not enough unless a normal motherboard
We love powerful 48-core processors, but not every customer is willing to pay for such capacity: even among modern companies with hybrid infrastructure (local + remote cloud), ...
Rapid test of the Palit GeForce RTX 2080 Super video card in machine learning tasks
If you choose a GPU not only for gaming, but also for scientific calculations, you are interested in seeing a video card with tensor cores and modern 8GB GDDR6 memory. Thanks to FP16 support, in some Teslas it shows a multiple speed advantage even over Tesla
Nvidia is planning to update its graphics cards, transferring the entire product line to the new GDDR6 memory. In our Geforce GTX 1660 Super review, we already said that in Machine Learning/Artificial Intelligence tasks, VRAM bandwidth is ...
