chapter1-AppendixA-m4-ziavras

Culler d patterson ucb latency simple operation wo

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: DDR SDRAM (4x,120x) • Disk: 3600, 5400, 7200, 10000, 15000 RPM (8x, 143x) Disk 10 (Latency improvement = Bandwidth improvement) 1 1 10 100 Relative Latency Improvement Some material is adapted from D. Culler & D. Patterson (UCB) (latency = simple operation w/o contention BW = best-case) S. Ziavras CPUs: Archaic vs. Modern • • • • • • • 1982 Intel 80286 12.5 MHz (80ns clock) 2 MIPS (peak) Latency 320 ns (instruct.) 134,000 xtors, 47 mm2 47 16-bit data bus, 68 pins Microcode interpreter, separate FPU chip • (no caches) Some material is adapted from D. Culler & D. Patterson (UCB) • • • • • • • 2001 Intel Pentium 4 1500 MHz (0.66ns) (120X) 4500 MIPS (peak) (2250X) Latency 15 ns (20X) 42,000,000 xtors, 217 mm2 217 64-bit data bus, 423 pins 3-way superscalar, Dynamic translate to RISC, Superpipelined (22 stage), Out-of-Order execution • On-chip 8KB Data caches, 96KB Instr. Trace cache, 256KB 256KB L2 cache S. Ziavras Latency Lags Bandwidth (last ~20 years) • Performance Milestones • Processor: ‘286, ‘386, ‘486, Pentium, Pentium Pro, Pentium 4 (21x,2250x) • Ethernet: 10Mb, 100Mb, 1000Mb, 10000 Mb/s (16x,1000x) • Memory Module: 16bit plain DRAM, Page Mode DRAM, Page 32b, 64b, SDRAM, DDR SDRAM (4x,120x) • Disk : 3600, 5400, 7200, 10000, 15000 RPM (8x, 143x) 10000 CPU high, Memory low Memory low (“Memory Wall”) 1000 Processor Network Relative Memory BW 100 Improve ment Disk 10 (Latency improvement = Bandwidth improvement) 1 1 10 100 Relative Latency Improvement Some material is adapted from D. Culler & D. Patterson (UCB) S. Ziavras Rule of Thumb Bandwidth improves by more than the square of the improvement in Latency th Some material is adapted from D. Culler & D. Patterson (UCB) S. Ziavras 6 Reasons Latency Lags Bandwidth 1. Moore’s Law helps BW more than latency • • Faster transistors, more transistors, more pins help Bandwidth » MPU Transistors: 0.130 vs. 42 M xtors (300X) » DRAM Transistors: 0.064 vs. 256 M xtors (4000X) » MPU Pi...
View Full Document

This document was uploaded on 02/09/2014.

Ask a homework question - tutors are online