CPU Core, Hyper-Threading, Clock Speed: Explained

CPU Core, Hyper-Threading, Clock Speed: Explained


So far, we have come across so many high-end
workstation specifications. 16 or 32 gigabytes of RAM, 4 terabytes of SSD storage, the jaw-dropping
teraflop ratings, ultimate graphics utilizing 8 gigabytes of memory, lightning speed processor
with 8 cores – or even 18 cores. But… what exactly is a core? A core is part of a CPU that receives instructions
and performs the computational part, or actions, based on those instructions as required by
a program. The set of instructions can allow a software
program to perform a specific function. Originally, CPUs had single core, which meant
that the CPU could handle only one task at a time since it had only one processing unit. To boost the performance, manufacturers implemented
dual cores in a single CPU. The processor is the entire chipset that is
composed of all the cores involved. Cores are like 2 or more independent parts
of the processor that do parallel processing, which means processing two different data
simultaneously in different units. This helps in multitasking without causing
much strain on the processor and affecting its performance. Each core can work on one task at a time,
so the more number of cores a CPU has, the more efficient and faster it is, since it
can perform multiple tasks at the same time. Or simply, it divides the workload into multiple
sections, then each core works on those sections and finally add their output together. A dual-core CPU literally has two cores, or
processing units, on the CPU chip. Like that, a quad-core CPU has four, an octa-core
CPU has eight, and so on. Using this concept of multiple cores improves
the performance significantly while maintaining a small CPU unit size. This also reduces the requirement of each
core to be cooled, supplied with power since they all share the same resources from the
parent CPU. Best of all, there is much less latency due
to faster communication between each other as a result of being on the same chipset. However, adding more cores to the CPU does
not come cheap. The overall price tag must be considered as
well. This is where hyper-threading comes to light. Manufacturers found hyper-threading as a substitute
for additional cores. A single physical CPU core with hyper-threading
appears as two logical CPUs to an operating system, while the CPU is still a single CPU. The actual CPU hardware can receive only a
single set of instructions for execution for each core, even though, with hyper-threading,
the operating system takes into account two CPUs for each core. It appears to the OS as if the CPU has double
the number of cores than it actually does, and it uses its own logic to speed up program
execution. In other words, it is just a clever scheduling
of data execution by the existing cores on the CPU. But this approach of hyper-threading is not
as much as efficient as having real extra cores on the CPU. Hyper-threading tends to be more effective
for multi-threaded workload than for single-threaded workload. Now let’s talk about clock speed of a CPU. Clock speed of a CPU is the speed or frequency
at which the CPU executes instructions of a program. It can be measured in megahertz (MHz) or gigahertz
(GHz), although you won’t hear “megahertz” a lot these days. It indicates how fast the PC fetches, decodes,
and executes the data. Hence, a higher frequency or clock speed means
better performance. CPUs have a range of clock speeds. It maintains a “base” speed when subjected
to low demanding tasks like playing music and sending emails or when the PC is just
idle. On the other hand, the CPU is able to reach
peak speed, or in fancy term “turbo speed”, when subjected to heavy tasks like gaming
or 4K video editing. This, however, as a side effect, generates
more heat and processors often “throttle” it to lower clock speeds to prevent overheating. So that’s all about it. Hope you guys liked the video, smash the like
button, don’t forget to subscribe and until next, stay tuned.

3 thoughts to “CPU Core, Hyper-Threading, Clock Speed: Explained”

  1. Very informative. I'm missing one bit of information thou.
    Is the speed per Thread 1/2 of the total core speed (f.i. does a 3.2 Ghz Core speed translate to 1.6 Ghz per thread)?
    It's hard to find this information online.

Leave a Reply

Your email address will not be published. Required fields are marked *