Modern enterprises rely on two broad categories of IT hardware: data center equipment for storing and computing data, and end-user devices like laptops. In this video, we’ll break down the four major components of data center equipment: servers, storage, networking, and ancillary devices. From understanding server types like tower, rack, and blade, to key performance metrics like processor chips and memory, we’ll give you a full primer on the hardware powering today’s IT environments.
▶️ Watch our other videos: • How Do We Make Old Power Grids Work with N...
Key Sections 0:24 - The 4 Distinct Groups under Data Center Equipment 1:08 - Servers Explained 1:37 - Larger Scale Servers at Data Centers 1:56 - Rack Servers Explained 2:25 - The Blade Server Explained 3:11 - The Metrics Behind Servers 3:35 - Performance of Servers
📚 This video serves as an excerpt from the Enterprise Hardware Industry Learning Primer, meticulously curated by the digital e-learning content provider, Primerli https://www.primerli.com/. The full module, comprising 10 bite-sized videos and quick quizzes, is tailored to empower professionals to grasp the industry intricacies in under an hour.
What industry do you want to learn about next? Let us know in the comments.
Enjoyed this video? Subscribe here: / @primerli
✅ Connect with us on LinkedIn: / primerli
Most modern enterprises use two broad types of IT hardware, so-called data center equipment
for storing and computing all that data, and end-user equipment like laptops and desktops.
We'll cover all of this. Within data center equipment, we'll cover four distinct groups.
Servers, or compute servers, which are for computing, doing calculations, and executing
code. Storage devices, which unsurprisingly are for data storage, long-term solutions for keeping
zeros and ones in place. Networking devices, which are what let these different devices
communicate with each other. And a final group we'll call ancillary data center equipment,
which supports the smooth and safe operations of the other hardware. Servers, storage devices,
and much of the ancillary equipment are sold by companies like Dell, HP, IBM, and Lenovo.
Networking devices, on the other hand, are a slightly different beast. The major companies
here include Cisco, Juniper Networks, and Huawei. Let's look into the first group, servers. We'll
cover the main types and the metrics commonly used to compare different products. First, the main
types of servers. One type is what's called a tower server. It looks like a large, old-fashioned
desktop PC. Tower servers are a more basic option, better suited for smaller businesses that are not
running full enterprise-scale data centers. The other two types of servers are designed
specifically for larger-scale setups, with more equipment. Data centers like these are
commonly space-constrained. Space is limited and expensive in terms of both real estate costs
and cooling costs. To maximize what is called data center density, companies want space-efficient
servers. The second type of server creates this space efficiency by being neatly stackable. This
is called a rack server because each one slots neatly into what are known as server racks. Racks
are like bookcases that house the servers, as well as other key hardware components. They're standard
fixtures, usually 19 inches wide. The biggest benefit of rack servers is flexibility. Each
works independently, with its own power source and cooling fans, making it relatively easy to
add or remove them. The third type of server is even more space-efficient, but less flexible.
This is called the blade server. Blade servers are not standalone units. They combine together
into a specialized box or chassis with a shared power and cooling system. This makes them more
space-efficient, but more difficult to add or remove. A bit confusingly, blade server chassis
will also slot neatly into that 19-inch wide server rack. Most of the major companies offer
a range of servers across types. For example, Dell's PowerEdge series, HP's ProLiant series,
and Lenovo's ThinkSystem series, each include tower, rack, and blade variants. Let's turn to
the common metrics people use to compare different server products. Of course, size is key. For rack
and blade servers, size is usually measured by vertical height, in terms of how many so-called
rack units they occupy. Each rack unit is roughly 1.75 inches. So, for example, a 2U server would
be roughly 3.5 inches tall. Performance is another key factor, usually gauged by the underlying
components, namely the number and type of semiconductor chips. All servers rely heavily
on chips. These are what manipulate digital zeros and ones during the computation itself.
Some of the most critical and advanced chips are called processors, including central processing
units, or CPUs, and graphics processing units, or GPUs. So-called memory chips are key as well,
like random access memory, or RAM, which provide the short-term memory needed during computation.
So, differentiating between servers is often about comparing the type and number of the CPUs used,
or the total RAM available, for example. And actually very similar chips and very similar
measures are used in end-user computers as well, but servers will simply use the most advanced
chips and many more of them. Lastly, we should say that even though this is a hardware primer,
software is key. Many servers have specific software for tasks like security or power
monitoring. These can all be part of the secret sauce that makes a server stand out.