The inner loop of a linear-equation solver executes the following operation: sumi+1 = sumi - axp(i)...
Fantastic news! We've Found the answer you've been seeking!
Question:
Transcribed Image Text:
The inner loop of a linear-equation solver executes the following operation: sumi+1 = sumi - axp(i) + b × q (i) Suppose that a, b, are constants and p(i), q(i) are array variables preloaded in Memory. All these data items can be streamed from Memory into an arithmetic pipeline, with one cycle delay between accesses to each data item, to avoid computation bottlenecks. We assume integer pipeline stages operating on integer data types. The objective is to perform N iterations, i = 1,2, ..., N of the above operation and to produce the final sum in minimal time. a. Design the block diagram and functional behavior of a three-function pipeline whose operations are Multiply, Add, and Subtract. Find the collision vectors for controlling the system and find the fastest possible cycle for the sequence of operations ×, +, and when operating on independent operands. Here, this does not account for the interlocking necessary to make sure that the value of sum used as an input is derived from the most recent value of sum used as an output. This means Memory Read and Write are considered functionally separate in their interactions with the pipe. b. Now consider the maximum speed attainable when the input to the Adder is interlocked (meaning Registered) to the output of the Subtractor. What is this maximum speed in your design? Note, a separate register is needed to achieve this task. C. If we want to produce one update of sum per cycle on the average, how can we modify the pipeline to achieve this rate? Show a pipeline architecture that achieves this rate and describe how you propose to control the pipeline. Discuss the timing issues without the actual hardware implementation. The inner loop of a linear-equation solver executes the following operation: sumi+1 = sumi - axp(i) + b × q (i) Suppose that a, b, are constants and p(i), q(i) are array variables preloaded in Memory. All these data items can be streamed from Memory into an arithmetic pipeline, with one cycle delay between accesses to each data item, to avoid computation bottlenecks. We assume integer pipeline stages operating on integer data types. The objective is to perform N iterations, i = 1,2, ..., N of the above operation and to produce the final sum in minimal time. a. Design the block diagram and functional behavior of a three-function pipeline whose operations are Multiply, Add, and Subtract. Find the collision vectors for controlling the system and find the fastest possible cycle for the sequence of operations ×, +, and when operating on independent operands. Here, this does not account for the interlocking necessary to make sure that the value of sum used as an input is derived from the most recent value of sum used as an output. This means Memory Read and Write are considered functionally separate in their interactions with the pipe. b. Now consider the maximum speed attainable when the input to the Adder is interlocked (meaning Registered) to the output of the Subtractor. What is this maximum speed in your design? Note, a separate register is needed to achieve this task. C. If we want to produce one update of sum per cycle on the average, how can we modify the pipeline to achieve this rate? Show a pipeline architecture that achieves this rate and describe how you propose to control the pipeline. Discuss the timing issues without the actual hardware implementation.
Expert Answer:
Answer rating: 100% (QA)
The question is asking to conceptualize a pipelined architecture for a linearequation solvers inner loop and optimize the computational efficiency of its pipeline Lets break this down into the three s... View the full answer
Related Book For
Probability And Statistics
ISBN: 9780321500465
4th Edition
Authors: Morris H. DeGroot, Mark J. Schervish
Posted Date:
Students also viewed these programming questions
-
Managing Scope Changes Case Study Scope changes on a project can occur regardless of how well the project is planned or executed. Scope changes can be the result of something that was omitted during...
-
The outer loop controls the number of students. Note that the inner loop of this program is always executed exactly three times, once for each day of the long weekend. Modify the code so that the...
-
Cancer is a frightening disease. The biological process that creates cancerous cells from healthy tissue is poorly understood at best. Much research has been conducted into how external conditions...
-
How are purchase discounts accounted for in a periodic inventory system?
-
What is the commonsense basis for expecting a union wage advantage? Explain how each of the following differences between union and nonunion firms might complicate ones determination of whether...
-
The 10 -ft wooden column has the dimensions shown. Determine the critical load if the bottom is fixed and the top is pinned. Take \(E_{\mathrm{w}}=1.6\left(10^{3} ight) \mathrm{ksi}, \sigma_{Y}=5...
-
Englehart Company sells two types of pumps. One is large and is for commercial use. The other is smaller and is used in residential swimming pools. The following inventory data is available for the...
-
Sketch Bohr's model for Mg (Magnesium). It is listed in the periodic table as: 24 12Mg (12 is the atomic number and 24 is the mass #). How many protons, electrons and neutrons are there?
-
Mr. M has been employed as an engineer by A Ltd., a company incorporated in Hong Kong. During the year ended 31 March 2019, Mr. M had the following income and expenditure. A monthly salary of...
-
Papaya, a specialty greeting card company, purchased a photocopier costing $48,000 on January 1, 2017. Installation costs were $2,000. The equipment is expected to produce a total of 4,500,000 copies...
-
10. 11. Bob and Lyn were unrelated partners in the BL Partnership, Bob owning a 70 percent interest and Lyn all the rest. Bob sold an old truck to the partnership for $3,000. The truck had been used...
-
A physicist has developed a device that generates a beam of alpha particles each of which is a helium nucleus, consisting of two protons, two neutrons, and no electrons. The device ejects a steady...
-
Suppose that the energy a wind turbine (WT) can produce before it fails for repair is exponentially distributed with an average value of 100 MWh (note that: IMWh-10 Watt-hour, for instance a 50W PC...
-
You calculated the ideal number of identical machines to assign to one operator as n' = 3.2 and the cost per operator-hour = $15; and the cost per machine-hour = $50. What is the correct numbers of...
-
Two point charges of magnitude +4.00 C and +2.00 C are placed at the opposite corners of a rectangle 1. 0.400 m, +4.00 C B +2.00 C, 0.800 m. a) What is the potential at point A due to these charges?...
-
To help financing its green projects, a UK based company is issuing a number of new bonds. These bonds have a face value of 1,000, 10% coupon rate and 5-year maturity. Assume that you can borrow from...
-
The packaging division of a company having considered several alternative package designs for the company's new product has finally brought down their choices to two designs of which only one has to...
-
For the conditions of Exercise 5, use the central limit theorem in Sec. 6.3 to find approximately the size of a random sample that must be taken in order that Pr(|n p| 0.1) 0.95 when p = 0.2.
-
The mode of a discrete distribution was defined in Exercise 12 of Sec. 5.2. Determine the mode or modes of the Poisson distribution with mean .
-
Consider again the three different conditions (a), (b), and (c) given in Exercise 2, but suppose now that p <1/2. For which of these three conditions is there the greatest probability that gambler A...
-
Find the fundamental frequency of longitudinal vibration of the tapered bar shown in Fig. 8.41, using Rayleigh's method with the mode shape \[U(x)=c_{1} \sin \frac{\pi x}{2 l}\] The mass per unit...
-
Approximate the fundamental frequency of a rectangular membrane supported along all the edges by using Rayleigh's method with \[W(x, y)=c_{1} x y(x-a)(y-b)\] \[V=\frac{P}{2}...
-
The root mean square value of a signal \(x(t), x_{\mathrm{rms}}\), is defined as \[x_{\mathrm{rms}}=\left\{\lim _{T ightarrow \infty} \frac{1}{T} \int_{0}^{T} x^{2}(t) d t ight\}^{1 / 2}\] Using this...
Study smarter with the SolutionInn App