28.4. Parallel Operations and Performance

So, in case of multiple bus systems there are multiple buses that connects the different components of the CPU that is obvious. So, there will be not a single data an address, bus there can be multiple buses to transfer the signals ok.

In a multiple bus system, the CPU utilizes more than one bus to connect its various components (such as the ALU, registers, and memory). This allows multiple signals to be transferred simultaneously, rather than having to wait for a single bus, which increases overall performance. Think of it like a multi-lane highway where cars can travel side by side, rather than a single-lane road where vehicles must take turns.

Imagine a team of workers in a factory. If they have multiple conveyor belts (buses), they can move different items to different places at the same time. However, if they only have one conveyor belt, they have to wait for one item to be delivered before the next one can start moving, causing delays.

Of course, one clear advantage as you can figure out if you have multiple buses to carry out data and control you require a much less control steps because many operations can be done in parallel.

Having multiple buses means that the CPU can perform several operations at the same time. With fewer steps needed to control the flow of data, tasks can be completed more quickly. This is analogous to multitasking, where a person can accomplish several tasks at once instead of doing them one after the other.

Consider a restaurant kitchen where several chefs work together. If each chef can work on different dishes simultaneously, the restaurant can serve meals much faster than if a single chef had to prepare each dish in sequence.

But again also we see some stray cases where not much advantage is there, but also at the same time you have to appreciate that it involves more cost.

While multiple buses offer performance increases, they also come with higher costs. More buses require more resources for manufacturing and controlling them, leading to increased complexity in design. Thus, a careful balance must be struck between performance and cost-efficiency.

Imagine a city that decides to build more roads to reduce traffic congestion. While more roads can improve traffic flow, they also require maintenance and increase budgetary expenditures. In some cases, it might be more efficient to optimize existing roads rather than build new ones.

Obviously, less number of control signals and temporary registers will be required if there are multiple buses as we will see that is quite obvious also, because if you have two lines you send the input and two lines to get the outputs you may not require any temporary registers because there will be two direct lines which will feed the ALU.

With multiple buses, the complexity of control signals and the need for temporary registers decreases. This is because direct connections allow for seamless data transfer to components like the ALU, removing bottlenecks that occur in single bus systems where temporary storage is often needed.

Think about how a team of organizers managing an event can streamline tasks with a communication setup. If organizers can communicate directly via walkie-talkies (multiple buses), they don’t need to write down messages (temporary registers) and pass them along, which saves time and improves efficiency.

So, program counter actually points to the current instruction, and then it will do program counter plus the next address. So, that it can program counter plus increment, which will point to the next address of the instruction so.

In a multi-bus architecture, the program counter (PC) can operate much more efficiently. Instead of needing multiple steps to update its value, the PC can directly compute and store the next instruction address in one step, thanks to multiple buses providing simultaneous data paths.

Imagine a librarian who can browse multiple bookshelves at once rather than looking at each shelf one by one. By being able to access different shelves simultaneously, the librarian can locate the next book much faster than if they had to do it sequentially.

The memory address register (MAR) tells that from which location of the memory data has to be brought in.

The memory address register (MAR) serves a critical role by pointing to the specific location in memory from which data needs to be fetched. In a multiple bus architecture, this function remains essential, but the potential for parallel operation means it could be used more efficiently if multiple memory sources existed.

Think of the MAR as a specific address on a delivery package. If a delivery person knows multiple addresses (representations of memory locations), they can fetch data from multiple locations more quickly than if they only had access to one address at a time.

Memory data register (MDR) will take some data from the memory and of course, you have distribute to some others places like, if you have an instruction called 𝑙𝑜𝑎𝑑 𝑅 ,𝑀.

The memory data register (MDR) holds data being transferred to and from memory. With multiple ports available in a bus architecture, data can be quickly distributed to several places, enhancing performance and speed of operations within the CPU.

Imagine a relay runner who hands off a baton to multiple teammates at once rather than sequentially passing it to just one. Each teammate can then continue the race simultaneously, showcasing the speed and efficiency of a well-coordinated team effort.

--