Scribd
Upload
20 views3 pages

PCA Written PartI

Uploaded by

MuhammadAsif
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
20 views3 pages

PCA Written PartI

Uploaded by

MuhammadAsif
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 3
Central Processing Unit | |__~ This model executes only |_one single instruction ey ! at one time, but _sperates only o a a | Few pieces of date Fetched /read from memery Von Neumann architecture Von Neumann bot tle neck b Separation of memory and CPU L Interconnect determing, rate which instructions ond data are accessible Process Entities | Executsble machine language proaram 2. Memory blocks (executable code, disks, nelwork interface cards) 3 Resource descriptors by the OS (file descriptor) 4. Steutiy occess ond levels (information on accessible hardware + software resources) 5: Process stole information (ready /biocked ete) Von Neumann models modification I Caching = Collection of memory lscations = Allows CPU te access more quickly then in main memory 2 Virtual Memory 7 To resolve the size limitation in caching = Main _memarg con act as cache for secondary storage = Creates more contiguous and lorger addresable memery space for applications & The Mechanism 1+ Address Space L Programs running ° computer has own virlual address space L Divides them into pages [in smaller units) L Can be larger than octual physical RAM 2- Page Table L Maps virtual address to LEnables system to find data / instruction that program needs eorres ponding physical addresses in RAM / disk 3- Page Swapping © Occurs when L Swaps data between RAM ond disk drive program tries to access date not in RAM. Advantages: I) Efficient Memory Usage 2) Implement isolation 3) Simplified Memory Monagement 3. Instruction level parallelism \+ Pipelining ~ Tasks overlapping between each other L Divides inshruchion to stages and allows mulli instructions $o be processed ot same ne L Each inshruction advances through stages in sequence 2- Mal tiple issue Simultaneously execute different instructions in a program © Runs multi instruction simultaneously in a single clock cycle L Minimizes dependencies » but requires more complex hardware Pipelining ewample: Instruction Pipelining in CPU (Stoge by Stage) Si - Instruction Fetch (iF) S1.- Instruction Decsde (10) S3- Execute (Ex) S4- Memory Access (MEM) SS- Write- Back (WB) Lv Malliple Tnsbruction: Superscalar Prscesing 1 Add RL R2, RF 2 Multiply R4, RS, Re 3. Subtmct RT, R3, RI 4 Division RIO, Ril, RI2 # Assume there are 2 mulbi issue processors, A) and A2: A [A T Me R+R | Ras Rs xRe Th Ry ~Ra-by | Po = Ru Qa

You might also like