lelClusterComputing2005TheTyrannyoftheStorage：LEL和集群计算2005暴虐的存储.ppt

OU Supercomputing Center for Education Research Parallel Cluster Computing 2005The Tyranny of the Storage Hierarchy Paul Gray, University of Northern Iowa David Joiner, Kean University Tom Murphy, Contra Costa College Henry Neeman, University of Oklahoma Charlie Peck, Earlham College Outline The Pathetic Progress of Computing Cast of characters What is the storage hierarchy? Registers Cache Main Memory (RAM) The Relationship Between RAM and Cache The Importance of Being Local Hard Disk Virtual Memory The Pathetic Progress of Computing The Pathetic Progress of Computing The Pathetic Progress of Computing The Pathetic Progress of Computing The Pathetic Progress of Computing So Says the Expert ... “CPU speed doubles every 18 months; compiler speed doubles every 18 years.” -- Stephen Wheat Director of HPC, Intel Cast of Characters Intel Pentium4 (including Xeon) Intel Itanium2 IBM POWER4 What is the Storage Hierarchy? Registers Cache memory Main memory (RAM) Hard disk Removable media (e.g., CDROM) Internet Henry’s Laptop Pentium 4 1.5 GHz w/1 MB L2 Cache 512 MB 400 MHz DDR SDRAM 40 GB 4200 RPM Hard Drive Floppy Drive DVD/CD-RW Drive 10/100 Mbps Ethernet 56 Kbps Phone Modem Storage Speed, Size, Cost Registers What Are Registers? Registers are memory-like locations inside the Central Processing Unit that hold data that are being used right now in operations. How Registers Are Used Every arithmetic or logical operation has one or more operands and one result. Operands are contained in registers. A “black box” of circuits performs the operation. The result goes into a register. How Many Registers? Typically, a CPU has less than 2 KB (2048 bytes) of registers, usually split into registers for holding integer values and registers for holding floating point (real) values, plus a few special purpose registers. Examples: IBM POWER4 (found in IBM pSeries supercomputers): 80 64-bit integer registers and 72 64-bit f