Performance and CPU Comparisons
SPEC Power Benchmark
As performance increases so does power consumption. Even at 0 performance, 80 watts are used
Clock cycles per instruction(CPI)
Average number of clock cycles per instruction for a program or program fragment. Clock Cycles= Instruction Count x Cycles per instruction CPU Time= Instruction Count x CPI, Clock Cycle Time CPU Time= (Instruction Count x CPI)/ Clock Rate
Relative Performance
Execution time(b)/Execution time(b) = n
ARMv7 is much closer to MIPS than ARMv8 since ARMv7 and MIPS came out at the same time.
False
CPU time depends on 3 factors: how many high-level language instructions an average CPI (clock cycles per instruction) the clock cycle time
False
The ARM corporation is a leading chip manufacturer.
False
The authors of our textbooks recommend using the MIPS (millions of instructions / second) metric for evaluating CPUs.
False
CPU time depends on 3 factors: how many machine language instructions an average CPI (clock cycles per instruction) the clock cycle time
True
Performance
1/Execution time
Instruction mix
A measure of the dynamic frequency of instructions across one or more programs
millions instruction per second(MIPS)
A measurement of program execution speed based on the number of millions of instructions. MIPS is computed as the instruction count divided by the product of the execution time and 10^6
Benchmark
A program selected for use in comparing computer performance.
General-purpose Register(GPR)
A register that can be used for addresses for data with virtually any instruction
Amdahl's Law
A rule stating that the performance enhancement possible with a given improvement is limited by the amount that the improved feature is used. It is a quantitative version of the law of diminishing returns T(improved)= (T(affected)/ improvement factor)+T(unaffected) T(unaffected) = (100-T(affected))
Workload
A set of programs run on a computer that is either the actual collection of applications run by a user or constructed from real programs to approximate such a mix. A typical workload specifies both the programs and the relative frequencies.
CPU Execution Time
Also called CPU time. The actual time the CPU spends computing for a specific task.
Throughput
Also called bandwidth. Another measure of performance, it is the number of tasks completed per unit time.
Response Time
Also called execution time, The total times required for the computer to complete a task, including disk accesses, memory accesses, I/O activities, operating system overhead, CPU execution time, and so on.
Clock Cycle
Also called tick, clock tick, clock period, clock, or cycle. The time for one clock period, usually of the processor clock, which runs at a constant rate
CPU Execution TIme for a Program
CPU clock cycles for a program x Clock cycle time, or CPU clock cycles for a program/ clock rate
ARMv8 is a 32-bit processor.
False(64 bit)
The x86 is a RISC architecture.
False(It's ARM and MIPS)
A CPU operating at 4 GHz has 4 clock cycles every nanosecond.
False(if it was Hz not GHz)
The fact that CPU manufacturers could not increasing the clock rate of uniprocessors, led to the development of: _________ __________ chips.
Multicore
Performance Ratio
Performance(A)/ Performance(B)
In the mid-2000s, CPU developers found it difficult to increase the clock rate of uniprocessors due to the
Power Wall
User CPU time
The CPU time spent in a program itself
System CPU Time
The CPU time spent in the operating system performing tasks on behalf of the program.
Clock Period
The length of each clock cycle
Instruction Count
The number of instructions executed by the program
ARMv8
● 2007 design began; released 2013 ● much closer to MIPS than ARMv7 because: ○ got rid of the 4-bit conditional execution field ○ has 32 general-purpose registers ○ includes a divide instruction
x86
● MIPS and ARM were developed by small groups, starting in 1985 ● x86 is the product of many groups over decades (look at evolution)
ARMv7
● chips used in phones, tablets ● ARM licenses their IP (intellectual property) to manufacturers ● ARM is a RISC architecture like MIPS ● MIPS has more registers ● ARM has more addressing modes and instruction formats