Assuming a computer has a single processor and a single core with no support for parallel execution, explain why running a multi-threaded program would show no performance improvement. Would inter-process communication be relevant in such a context?
Answer)
When a computer has a single processor and is single cored then while running the multi thread program context switching would be needed but when the scheduler would choose the upcoming process to be executed it will have to check for the current register's so as to backup the register's state and also in the upcoming process would also need the current state or switch pages or load additional data's which is not feasible when one processor and single core. Hence the efficiency would not be improved but same or even less due to load.
Inter process communication with time slicing or scheduling can help to some extent in some scenario's.
Hope this answer helps. :) Happy to help.
Assuming a computer has a single processor and a single core with no support for parallel...
Assume for arithmetic, load/store, and branch instructions, a processor core has CPIs of 1,12 and 5 respectively. Also assume that on a single processor core a program requires the execution of 2.56E9 arithmetic instructions, 1.28E9 load/store instructions, and 256.0E6 branch instructions. Assume all processor cores run at 2 GigaHertz. Assume that, as the program is parallelized to run over multiple processor cores, the number of load/store instructions per processor core is divided by .5 times p for all number of...
Consider the speedup gained by a multicore processor over a single processor. Use Amdahl’s law to calculate the speedup for the hypothetical scenarios below. Assume the total execution time of the program is 10 seconds. Assume N is the number of parallel processors. Assume 1-f and f are the execution time (fraction) involving code that is sequential and parallelized, respectively. Show your work and explain the size of the speedup (e.g., none, small, large, etc.). N=1, f=1. N=2, f=0.5 N=3,...
Assume for arithmetic, load/store, and branch instructions, a processor has CPls of 3, 15, and 7, respectively. Also assume that on a single processor a program requires the execution of 3.12E9 arithmetic instructions, 1.75E9 load/store instructions, and 512Million branch instructions. Assume that each processor has a 2.5GHz clock frequency. Assume that, as the program is parallelized to run over multiple cores, the number of arithmetic and load/store instructions per processor divided by 0.8xp ( where p is the number of...
Suppose we have developed new versions of a processor with the following characteristics: Version Voltage 1.5 V 1.2V Clock rate 2.4 GHz 3GHZ a) How much has the capacitive load varied between versions if the dynamic power has been reduced by 15%? Assuming that the capacitive load of version 2 is 70% the capacitive load of version 1, find the new voltage value for version 2 if the dynamic power of version 2 is reduced by 35% from version 1....
Using the Critical Thinking Process, identify 5 disk scheduling options, and 5 disk scheduling criteria. Weigh the options against the criteria and select the best disk scheduling option for: Identify and select a process-to-process communication technique that best both upward and downward scalability for a non-real time mission-critical system your company has been building. In This Format: An approach to making informed, unemotional decisions to is employ the following 4 step process: A) Identify your options B) Establish your criteria...
A single NEC SX-9 vector processor unit (VPU) has a peak arithmetic performance of 102.4 GFs/s. The memory system supplies a peak memory bandwidth of 162 GBs/s. The VPU dissipates 40 W, while the memory system dissipates 55 W. Together, both components dissipate 95 W. Every program has an arithmetic intensity (AI), measured in flops/byte, that indicates how much performance is achievable for a given bandwidth. The equation is: achievable GFs/s = min{AI*BW,peak GFs/s}. a) What is the minimum AI...
QUESTION 1 . ______________ allow(s) a computer to invoke procedures that use resources on another computer Pervasive computing Remote procedure calls (RPCs) Cloud computing Global computing QUESTION 2 The simplest example of a neural net is the: CPu perceptron systolic array supervised learning network QUESTION 3 The first company in the world to manufacture and sell what it identifies as a quantum computer is: D-Wave Computers Cray Google Intel QUESTION 4 A ______________ is a collection of distributed workstations that...
4. A multi-programmed operating system running on a single CPU assigns time slices of CPU time to the virtual machines that run the individual programs. Each virtual machine is given access to the CPU for a certain time (the time slice), during which it will continue to execute its program. The next time slice then goes to another virtual machine, and so on. Assume that each time slice is one millisecond (1/1000 of a second) long and the computer on...
1. Given the following instruction sequence for the MIPS processor with the standard 5 stage pipeline $10, S0. 4 addi lw S2.0(S10) add sw S2,4(510) $2, $2, $2 Show the data dependences between the instructions above by drawing arrows between dependent instructions (only show true/data dependencies). a. Assuming forwarding support, in what cycle would the store instruction write back to memory? Show the cycle by cycle execution of the instructions as they execute in the pipeline. Also, show any stalls...
4. A multi-programmed operating system running on a single CPU assigns time slices of CPU time to the virtual machines that run the individual programs. Each virtual machine is given access to the CPU for a certain time (the time slice), during which it will continue to execute its program. The next time slice then goes to another virtual machine, and so on. a. Assume that each time slice is one millisecond (1/1000 of a second) long. The computer on...