Computer Organization & Architecture – Key Microprocessor Specs.pptx
- 1. In the Name of Allah Who Is the Most Compassionate and the Most Merciful
- 2. • COA Presentation • Lecturer: P. Rahmatzai • Group :Azizullah Yosufi, Abdul Khalil Mohammadi, Jawad Rahimi, Rooheen Sayeedi & Najeebullah Karimi • Semester 4, May 2024
- 4. Contents ★ Clock Speed • Clock, Clock Cycle, IPC, IPS, Overclocking, Underclocking ★ Bus Width • Data Bus, Address Bus, Memory Bus ★ Transistors ★ Feature Size ★ Addressable Memory • Address Space ★ Virtual Memory • Paging, Page Faults, MMU ★ Cache Memory • Cache Levels #0
- 5. Clock Speed The CPU clock speed is a measure of how fast a computer’s central processing unit (CPU) can execute instructions. It’s like the rhythm at which the CPU operates, and it’s measured in gigahertz (GHz), which is a billion cycles per second. To better understand clock speed, let’s first take a look at the CPU clock & clock cycle. • CPU Clock: The CPU clock is an internal timing mechanism within the CPU that generates a series of electrical pulses at regular intervals. These pulses are used to synchronize the operations of the CPU, ensuring that all its parts work together in a coordinated fashion; it’s like a conductor in an orchestra, making sure every musician plays in time with the others. #1
- 6. • Clock Cycle: Each electric pulse generated by the CPU clock corresponds to one clock cycle. During each clock cycle, the CPU can perform a basic operation, such as executing an instruction or accessing memory. • Clock Speed: This is the frequency at which the clock cycles occur. A higher clock speed means more cycles per second, allowing the CPU to execute more instructions in the same amount of time. • Gigahertz (GHz): This is the unit of measurement for clock speed. One GHz equals one billion cycles per second. So, a 3 GHz CPU can perform three billion cycles each second. • Better Performance: A higher clock speed means the CPU can execute more instructions per second, leading to faster performance. This is especially true for single-threaded tasks that rely on a single CPU core. #2 Clock Speed (cont.)
- 7. • Instructions Per Clock Cycle (IPC): IPC measures the average number of instructions a CPU can execute in one clock cycle. An indicator of a CPU's efficiency, a higher IPC means the CPU is more efficient, getting more done in each cycle. For example, if a CPU has an IPC of 2, it can execute two instructions for every clock cycle. • Instructions Per Second (IPS): IPS is a measure of the total number of instructions a CPU can execute in one second. It’s calculated by multiplying the CPU’s clock speed by its IPC. So, if a CPU has a clock speed of 3 GHz & an IPC of 2, it can execute 6 billion instructions per second. In essence, while IPC is about the efficiency of a CPU at a micro level (per cycle), IPS is about the overall performance at a macro level (per second). #3 Clock Speed (cont.)
- 8. • Base and Turbo Speed: CPUs have a base clock speed, which is the guaranteed minimum speed under normal conditions. When more processing power is needed, the CPU can boost its speed temporarily to a higher turbo speed. • Power Saving: To save energy & reduce heat, CPUs can lower their clock speed when the system is idle or under light use. This is similar to a car idling at a stoplight to conserve fuel. • Thermal Management: If the CPU gets too hot, it may throttle its speed to cool down and prevent damage. This is like a runner slowing down to catch their breath. #4 Clock Speed (cont.)
- 9. • Overclocking: This is the process of manually increasing the clock rate of a computer’s CPU beyond the manufacturer’s specified speed. The main benefit of this approach is improved performance; however, without advanced cooling solutions, it can lead to additional heat and more power consumption. It’s worth mentioning that not all CPUs support overclocking. • Underclocking: The opposite of overclocking, also known as downclocking, it refers to the process of configuring a CPU to operate at a lower clock speed than its rated max. Commonly used in scenarios where maximum performance isn’t necessary, this can lead to power efficiency, less heat, reduced noise and greater stability. While the clock speed is a good measurement of performance, it’s not the only factor to consider. Other elements such as core count, bus speed, hard drive, RAM, and SSD also play significant roles in determining a computer’s overall speed and efficiency. #5 Clock Speed (cont.)
- 10. When we’re generally talking about the bus width, we usually mean the width of the data bus; nonetheless, let’s take a quick look at the three types of buses making up the system bus or omnibus: Data bus, Address bus & Control bus. • Data Bus: The data bus is responsible for transferring data & instructions between the CPU & other parts of the computer, like memory or I/O, making it a bidirectional pathway. The width of the data bus, measured in bits, indicates how much data can be transferred at one time. For example, a 32-bit data bus can transfer 32 bits of data in a single operation. The wider the data bus, the more data it can move simultaneously, which can lead to significant performance improvements. That is why 64-bit data buses are common in modern-day CPUs. #6 Bus Width
- 11. • Address Bus: The address bus carries the specific addresses of memory locations where data is stored. When the CPU needs to read from or write to a particular location in memory, it sends the address of that specific location over the address bus to the memory or I/O; therefore, unlike data bus, this bus is unidirectional. The width of the address bus, also measured in bits, determines the maximum amount of memory the CPU can address. For example, a 32-bit address bus can address up to 232 memory locations, which equals 4 gigabytes (GB) of memory space. It’s also commonplace to be 64 or 48 bits nowadays, theoretically allowing the CPU to address way more memory than actually present in the system. #7 Bus Width (cont.)
- 12. • Control Bus: The control bus carries control signals from the CPU to various parts of the computer, including memory, I/O devices & secondary storage. These signals are essential for orchestrating the activities of the computer’s components and include Read/Write Signals, Interrupt Requests, Status Lines, etc. Control Bus is also bidirectional, meaning it can send signals in both directions. The width of the control bus is mostly expressed in terms of the number of different lines it has for facilitating the transfer of diverse types of control signals. #8 Bus Width (cont.)
- 13. A transistor is a semiconductor device that can amplify or switch electronic signals and electrical power. It is an essential component in virtually all electronic devices, acting as the fundamental building block of microchips, including the CPU of a computer. Transistors are responsible for creating the binary 0’s and 1’s (bits) that computers utilize to communicate and process information using Boolean logic. They can be combined to create logic gates, which are the building blocks of digital circuits. Ever since the second generation of computers, transistors have been the cornerstone of computing getting increasingly integrated over time & giving birth to today’s powerful devices — alongside other technological advances in both hardware & software. #9 Transistors
- 14. The feature size for a microprocessor essentially refers to the size of its individual transistors. More specifically, it often relates to the smallest part of the transistor that can be manufactured, such as the gate length, which controls the flow of electricity within the transistor. Typically measured in nm (nanometers), this size is crucial because it impacts the overall density, speed and efficiency of the microprocessor. Smaller feature sizes allow for more transistors to fit on a chip, which can lead to more powerful and energy-efficient processors. The industry is now producing chips with feature sizes as small as 5 or 3 nm and the trend is going on in full swing. In short, the reduction in feature size is a key aspect of the technological advancements we observe today. #10 Feature Size
- 15. The addressable memory for a microprocessor is the total amount of memory that the processor can use or “address” directly through its instructions. As we know that the memory in a computer is organized into cells, each cell has a unique address & the microprocessor uses these addresses to access specific memory cells. • Address Space: The address space’ size, which determines how much memory can be addressed, depends on the width of the address bus. For instance, as stated before, a 32-bit address bus can address 232 unique addresses, which equates to 4 GB of addressable memory; this means that even if a computer has 16 GB of RAM onboard, due to the small address bus, it’ll only be able to access 4 GB of it. That’s why 64-bit buses ensure not only full access but also futureproofing; however, operating systems, hardware & processor architectures do impose a limitation on the theoretical maximum of 16 EB (exabytes). #11 Addressable Memory
- 16. Virtual memory is a memory management capability of an operating system that uses hardware and software to allow a computer to compensate for physical memory shortages, temporarily transferring data from random access memory (RAM) to disk storage. This process creates the illusion to users and applications that there is more RAM available than is physically present. • Memory Extension: Virtual memory extends the available memory on a system by using a portion of the hard drive (called the paging file) to act as additional RAM. • Virtual Address Generation: When a program is running, it generates virtual addresses. These addresses are used within the program’s own virtual address space & do not correspond directly to physical memory locations. The separation between processes is vital for memory protection & multitasking. #12 Virtual Memory
- 17. • Page System: The OS divides each process into pages and keeps their collection in a data structure called a page table. Each entry in the page table corresponds to a page of memory, which is typically 4 KB in size. Aside from collection, the table also contains mappings from the virtual addresses (pages) to physical addresses (frames in the RAM). • Memory Management Unit (MMU): The CPU includes a component called the MMU which is responsible for translating virtual addresses into physical addresses. When a program accesses a memory location, the MMU looks up the virtual address in the page table to find the related physical address. • Swapping: If there is not enough physical memory to hold all the needed pages, the operating system may swap some pages to disk, freeing up physical memory. When those pages are needed again, they are swapped back into RAM, and the page table is updated accordingly. #13 Virtual Memory (cont.)
- 18. • Page Faults: If the MMU cannot find a virtual address in the page table, this causes a page fault. The operating system then handles the page fault by loading the required page into physical memory and updating the page table with the new mapping. Different algorithms are used to decide which pages to swap in order to make space for the new pages. • Size: The amount of virtual memory available in a system is not fixed by the processor itself but is determined by the OS and the system’s configuration. The initial size of the paging file to be 1.5 times the size of the physical RAM and the maximum size to be about 3 times the size of the RAM 4. So, for example, if a computer has 8 GB of RAM, the virtual memory can be configured to be between 12 GB to 24 GB. However, these values can be adjusted manually for specific needs or system configurations. #14 Virtual Memory (cont.)
- 19. Cache memory is a special type of fast, small memory in a computer that serves as a buffer between the CPU & the main memory (RAM). It’s designed to speed up the process of data retrieval by storing frequently used data and instructions close to the CPU, drastically improving performance. ★ How Cache Memory Works: • Data Retrieval: When the CPU needs data, it first checks the cache memory. If the data is there (a “cache hit”), it can be used immediately. If not (a “cache miss”), the data is fetched from the main memory and also stored in the cache for future access. • Locality of Reference: Cache memory exploits the principle that programs tend to use a relatively small portion of their address space at any given time, and they often need the same data and data near it repeatedly. #15 Cache Memory
- 20. ★ Cache levels: This refers to the hierarchy of cache memory within a computer’s CPU. Modern CPUs have multiple levels of cache, each designed to improve the efficiency and speed of data access for the processor. • Level 1 Cache (L1): L1 cache is the smallest and fastest type of cache memory, located directly on the CPU chip. It provides the CPU with immediate access to the most frequently used data & instructions and stores the most critical pieces of data. Its size typically ranges from 2KB to 64KB. • Level 2 Cache (L2): L2 cache is larger than L1 but still very fast. It may be located on the CPU chip or on a separate chip close to the CPU. Acts as a buffer between the L1 cache and the slower main memory (RAM). Usually ranges from 256KB to 2MB. #16 Cache Memory (cont.)
- 21. • Level 3 Cache (L3): L3 cache is larger and slower than L1 and L2, but still much faster than main memory. It’s often shared among multiple CPU cores & stores data that is less frequently accessed but still needs to be quickly available. It can range from 4MB to 50MB or more, depending on the CPU design. In summary, cache levels are a structured approach to storing and accessing data in a way that maximizes speed and efficiency for the CPU. Each level serves a specific purpose in ensuring that the processor has the quickest possible access to the data it needs to perform computations effectively. #17 Cache Memory (cont.)
- 23. • Microsoft Copilot • Template Design by SmileTemplates.com © Sources #19
- 24. Thanks a lot; I’m very grateful!