Fill in all blank spaces. All miss penalties should be rounded to the nearest integer. All CPI values should be rounded to one decimal place, if needed. Suppose the following is true about a CPU: It has only L1 cache . The base CPI is 1.5 The clock rate is 2.5 GHz . The memory access time is 250 ns . The miss rate per instruction at L1 cache is 4% The cache miss penalty is cycles. The new effective CPI is Now suppose the following about the same CPU: . It also has L2 cache • Access time for L2 cache is 13 ns . The miss rate to main memory is reduced by 0.3% The miss penalty for an L1 miss with an L2 hit is The miss penalty for an L1 miss with an L2 miss is The new total CPI in this case is cycles. cycles.

Fill in all blank spaces. All miss penalties should be rounded to the nearest integer. All CPI values should be rounded to one decimal place, if needed. Suppose the following is true about a CPU: It has only L1 cache . The base CPI is 1.5 The clock rate is 2.5 GHz . The memory access time is 250 ns . The miss rate per instruction at L1 cache is 4% The cache miss penalty is cycles. The new effective CPI is Now suppose the following about the same CPU: . It also has L2 cache • Access time for L2 cache is 13 ns . The miss rate to main memory is reduced by 0.3% The miss penalty for an L1 miss with an L2 hit is The miss penalty for an L1 miss with an L2 miss is The new total CPI in this case is cycles. cycles.

C++ for Engineers and Scientists

4th Edition

ISBN:9781133187844

Author:Bronson, Gary J.

Publisher:Bronson, Gary J.

Chapter2: Problem Solving Using C++using

Section2.3: Data Types

Problem 9E: (Practice) Although the total number of bytes varies from computer to computer, memory sizes of...

See similar textbooks

Similar questions

The CPU clock is 1 GHz. One OP requires 5 clock cycles (for example arithmetic instruction and conditional). Each memory access (read or write) requires 100 clock cycles and the problem size is 1 million or n = 1,000,000. There is no memory access cost of loop index variable: don't count any arithmetic instructions for initializing or incrementing the loop index value. Assume the loop index variable is free of charge. There is an arithmetic operation involved in updating the accumulator variable. Algorithm Sum = 0 For i in range (0, n); Sum += i; Print (“sum =, sum) How many memory accesses and clocks are required by this algorithm?
The CPU clock is 1 GHz. One OP requires 5 clock cycles (for example arithmetic instruction and conditional). Each memory access (read or write) requires 100 clock cycles and the problem size is 1 million or n = 1,000,000.There is no memory access cost of loop index variable: don't count any arithmetic instructions for initializing or incrementing the loop index value. Assume the loop index variable is free of charge. There is an arithmetic operation involved in updating the accumulator variable. AlgorithmSum = 0A = list (range(0,N)) //vectorFor i in range (0, n);Sum += A[i];Print (“sum =, sum) In this algorithm how many memory accesses and clocks are required?
The memory access time is I nanosecond for read operation with a hit in cache. * nanoseconds for a read operation with a miss in cache, 2 nanoseconds for a write operation with a hit in cache and 10 anoseconds for a write operation with a miss in cache. Execution of a sequence of instructions involves 100 instruction fetch operations, 60 memory operand read operations and 40 memory operand write operations. The cache hit-ratio is 0.9. The average memory access time (in nanoseconds) in executing the sequence of instructions is
There are many parameters that could be used to describe disk performance; among them are: number of bits per track disk capacity (in bits) number of disk surfaces rotational speed rotational latency transfer rate tracks per surface sectors per track blocks per track sectors per block seek time speed of disk arm block-read time number of blocks Some of these parameters are independent, and others are (approximately) linearly related. That is, doubling one doubles the other. Decide which of these parameters are linearly related. Then, select from the list below, the relationship that is true, to within a close approximation. Note: none of the statements may be true exactly, but one will always be much closer to the truth than the other three. Also note: you should assume all dimensions and parameters of the disk are unchanged except for the ones mentioned. a) If you double the number of sectors per block, then you double the number of blocks on the disk. b)…
There are many parameters that could be used to describe disk performance; among them are: number of bits per track disk capacity (in bits) number of disk surfaces rotational speed rotational latency transfer rate tracks per surface sectors per track blocks per track sectors per block seek time speed of disk arm block-read time number of blocks Some of these parameters are independent, and others are (approximately) linearly related. That is, doubling one doubles the other. Decide which of these parameters are linearly related. Then, select from the list below, the relationship that is true, to within a close approximation. Note: none of the statements may be true exactly, but one will always be much closer to the truth than the other three. Also note: you should assume all dimensions and parameters of the disk are unchanged except for the ones mentioned. a) If you divide tracks into half as many blocks, then you double the read time for a block.…
funCount: MOVLW .15 MOVWF Ox82 count: DECF Ox82 BNZ count RETURN Assuming Oscillator frequency is set to 4 Mhz, fill in the following blanks based on the execution of the above code segment: Instruction cycle period, Tinst = uSec Number of instruction cycles required to execute the code Total time to execute the code uSec
Assume a program has 20% store and load. When we run this program in a computer, we observe that L1 instruction cache has a hit rate of 98% and L1 data cache has a hit rate of 96%. The L2 cache has a hit rate of 80%. The access time of L1 cache is 2 cycle, and the access time of L2 cache is 40 cycles. If a miss occurs in L2, we have to go to main memory and the penalty is 100 cycles. Assume that the cache uses physical address.a) Without a TLB, what is the AMAT?b) Assume a TLB is built in the cache and its access time is ignored. If the TLB has a 99% hit rate, what is the new AMAT?c) If there is a 0.01% chance that a page fault happens, and the penalty to fetch a page from the disk is 106cycles, what is the new AMAT?
Consider a fully-associative cache of size 4. Each slot in the cache can have just one item (i.e. the line size is 1 item). The cache is empty to start with. The cache uses an LRU replacement policy: every slot has a counter; every time a slot is accessed, a global counter is incremented and the value is stored in the slot counter; the slot with the lowest counter value is chosen for replacement. Sequence Id 1 2 3 4 5 6 7 8 10 Address Ox0012 0x0014 Ox0016 Ox0018 0x0016 0x0012 0x0020 Ox0022 0x0014 Ox0012 Hit/Miss Accesses 1 to 10 are respectively: Select one: O a. Miss, Miss, Miss, Miss, Hit, Hit, Miss, Miss, Miss, Miss O b. Miss, Miss, Miss, Miss, Hit, Hit, Miss, Miss, Hit, Hit O. Miss, Miss, Miss, Miss, Hit, Hit, Miss, Miss, Hit, Miss O d. Miss, Miss, Miss, Miss, Hit, Miss, Miss, Miss, Miss, Hit O e. Miss, Miss, Miss, Miss, Hit, Hit, Miss, Miss, Miss, Hit
A cache memory has a line size of eight 64-bit words and a capacity of 4K words. The main memory size that is cacheable is 1024 Mbits. Assuming 4-way set associative mapping and that the addressing is done at the byte level. What is the format of the main memory addresses (i.e s-d, d, and w)? For the hexadecimal main memory location 2BFACEDH, find the corresponding 4-way set-associative memory format
instruction is in the first picture cacheSim.h #include<stdlib.h>#include<stdio.h>#define DRAM_SIZE 1048576typedef struct cb_struct {unsigned char data[16]; // One cache block is 16 bytes.u_int32_t tag;u_int32_t timeStamp; /// This is used to determine what to evict. You can update the timestamp using cycles.}cacheBlock;typedef struct access {int readWrite; // 0 for read, 1 for writeu_int32_t address;u_int32_t data; // If this is a read access, value here is 0}cacheAccess;// This is our dummy DRAM. You can initialize this in anyway you want to test.unsigned char * DRAM;cacheBlock L1_cache[2][2]; // Our 2-way, 64 byte cachecacheBlock L2_cache[4][4]; // Our 4-way, 256 byte cache// Trace points to a series of cache accesses.FILE *trace;long cycles;void init_DRAM();// This function print the content of the cache in the following format for an N-way cache with M Sets// Set 0 : CB1 | CB2 | CB 3 | ... | CB N// Set 1 : CB1 | CB2 | CB 3 | ... | CB N// ...// Set M-1 : CB1 | CB2 | CB…
A program is profiled on a CPU: 20% of all instructions are arithmetic, 40% are load, 20% are store, 20% are conditional branches. On this CPU, loads take 4 cycles, stores 1 cycle, ALU instructions 2 cycle, and branches 3 cycles. Compute the CPI for each instruction and the average CPI for that program on that CPU: Compute the CPI for each instruction (GIVE ANSWERS WITH 2 DECIMAL DIGITS: CPI for loads is CPI for stores is CPI for ALU is CPI for branches is Average CPI is
Q1. Compute the effective CPI for an implementation of a RISC-V CPU using The Figurel. Assume we have made the following measurements of average CPI for instruction types: Instruction All ALU operations Loads Stores Branches Jumps Program astar bzip gec gobmk h264ref hmmer libquantum mcf omnetpp perlbench sjeng xalancbmk Loads 28% 20% 17% 21% 33% 28% 16% 35% 23% 25% 19% 30% Stores 6% 7% 23% 12% 14% 9% 6% 11% 15% 14% 7% 8% Clock cycles 1.1 4.2 3.3 2.8 3.0 Branches 18% 11% 20% 14% 5% 17% 29% 24% 17% 15% 15% 27% Jumps 2% 1% 4% 2% 2% 0% 0% 1% 7% 7% 3% 3% ALU operations 46% 54% 36% 50% 45% 46% 48% 29% 31% 39% 56% 31% Average the instruction frequencies of astar and perlbench to obtain the instruction mix.