DerekChiou_EE360N_Spring2010_Lecture5

DerekChiou_EE360N_Spring2010_Lecture5 - Lecture 5: Memory...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
© Derek Chiou Lecture 5: Memory Prof. Derek Chiou University of Texas at Austin
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Test of size © Derek Chiou: EE360N: Lecture 5 2 Recap Microcoded machines and memory Alternate ISA structures Simple pipelining A simple pipelined machine is an “unrolled” microcoded machine Outline Finish pipelining Memory SRAM/DRAM 09/28/10
Background image of page 2
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 3 Why Pipeline? No wasted resources, no need to pipeline Share Resources Register file was sitting around doing nothing anyways Instead of an instruction consuming all of processor, consumes only part of processor (between pipeline registers) Increase Issue Rate Issue 1 instruction per cycle, instead of one every other cycle Increase overall performance Potentially increase clock rate Examples Track (wave pipeline) Disneyland (Space Mountain verses Star Tours) Car wash
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 4 Definitions Instructions Per Clock Cycle (IPC) Average number of instructions executed per clock cycle Clock Cycles per Instruction (CPI) Average number of clock cycles per instruction Inverse of each other Dependent on application, ISA, micro-architecture, system DO PROBLEM 6
Background image of page 4
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 5 Why Is Clock Rate Important? ClockCycle Time Inst s ClockCycle NumInst TotalTime * * =
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 6 Why Is Clock Rate Important? ClockCycle Time Inst s ClockCycle NumInst TotalTime * * = CycleTime CPI NumInst TotalTime * * =
Background image of page 6
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 7 Pipelining is Good, Can We Pipeline More? If ALU pipelined, can we issue twice as fast again?
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Test of size 09/28/10 © Derek Chiou: EE360N: Lecture 5 8 Pipelining is Good, Can We Pipeline More? If ALU pipelined, can we issue twice as fast again?
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 32

DerekChiou_EE360N_Spring2010_Lecture5 - Lecture 5: Memory...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online