Question

Hi can you help me solve the following Computer Architecture and Organisation Problem on (Performance: Measurement and Analysis)? I have provided some suggested guidelines in answering the questions. Thank you.

1. What is throughput? Will a system with higher throughput be faster than a system with lower throughput? Please provide your analysis with clearly stated examples. (Difference between throughput and response time) Ans: Not faster. Why? 2. What does BPS (bits per second) mean in term of performance? (i.e Throughput or Response Time) 3. Amdahl law is a law of deminish return. Please explain Amdahl Law. (Is it good or bad? How does it help?)

Answer 1

1)Throughput: Throughput can be defined as number of processes completed per unit time (or) It can be defines as the amount of work that can be performed in a given period of a time.

Example related to supply chain management

A company’s throughput also depends on how well the firm manages its supply chain, which is the interaction between the company and its suppliers. Assume, for example, that ABC Cycles manufactures bicycles. The firm has procedures in place to maintain equipment used to make bikes, and it plans production capacity based on scheduled machine maintenance and employee staffing plans. However, ABC also needs to communicate with its suppliers for metal bike frames and seats, because the suppliers need to deliver components parts when ABC needs them for production. If the parts don't arrive when ABC Cycles needs them, ABC's throughput will be lower.

Difference between Response time and throughput

Response time : Response time, in the context of computer technology, is the elapsed time between an inquiry on a system and the response to that inquiry. Used as a measurement of system performance, response time may refer to service requests in a variety of technologies. Low response times may be critical to successful computing.

Throughput : Throughput can be defined as number of processes completed per unit time (or) It can be defines as the amount of work that can be performed in a given period of a time.

2)BPS : Bits per second is a technology term, Which is standard way to measure data transfer rates, such as network connection and internet download speed (or) A benchmark can be used to measure throughput. In data transmission, network throughput is the amount of data moved successfully from one place to another in a given time period, and typically measured in bits per second (bps), as in megabits per second (Mbps) or gigabits per second (Gbps).

3)Amdhal law :  It is a formula which gives the theoretical speedup in latency of the execution of a task at a fixed workload that can be expected of a system whose resources are improved. (or)  It is a formula used to find the maximum improvement possible by just improving a particular part of a system. It is often used in parallel computing to predict the theoretical speedup when using multiple processors.

Mathematically, Amdahl's law is formulated as follows:

Throughput: Throughput can be defined as number of processes completed per unit time (or) It can be defines as the amount of work that can be performed in a given period of a time. Example related to supply chain management A company's throughput also depends on how well the firm manages its supply chain, which is the interaction between the company and its suppliers. Assume, for example, that ABC Cycles manufactures bicycles. The firm has procedures in place to maintain equipment used to make bikes, and it plans production capacity based on scheduled machine maintenance and employee staffing plans. However, ABC also needs to communicate with its suppliers for metal bike frames and seats, because the suppliers need to deliver components parts when ABC needs them for production. If the parts don't arrive when ABC Cycles needs them, ABC's throughput will be lower. Difference between Response time and throughput Response time: Response time, in

Where

Smax is the maximum possible improvement of the system

p is the part that can be improved. In other words, (1-p) is the part of the system that cannot be improved.

s is the performance improvement factor of p after applying the enhancements.

Let us now look at how the maximum improvement can be calculated using Amdahl's law. For a given system, if the part that can be improved is 25% of the overall system and its performance can be doubled, then

Relating to law of deminish return :

Amdahl's law is often conflated with the law of diminishing returns, whereas only a special case of applying Amdahl's law demonstrates law of diminishing returns. If one picks optimally (in terms of the achieved speedup) what to improve, then one will see monotonically decreasing improvements as one improves. If, however, one picks non-optimally, after improving a sub-optimal component and moving on to improve a more optimal component, one can see an increase in the return. Note that it is often rational to improve a system in an order that is "non-optimal" in this sense, given that some improvements are more difficult or require larger development time than others.

Amdahl's law does represent the law of diminishing returns if on considering what sort of return one gets by adding more processors to a machine, if one is running a fixed-size computation that will use all available processors to their capacity. Each new processor added to the system will add less usable power than the previous one. Each time one doubles the number of processors the speedup ratio will diminish, as the total throughput heads toward the limit of 1/(1 − p).

Benfits of Amdahl's law :

1)Uses in memory hieararchy design

2)use in Instruction set and processor design

In Amdahl’s law, computational workload W is fixed while the number of processors that can work on W can be increased. Where as in Gustafson's law that increase of problem size for large machines can retain scalability with respect to the number of processors.

Amdahl's law is not quite good compare to Gustafson's law,because of its fixed size.