30 Questions
What is the main characteristic of the instruction set in a RISC architecture?
Few and simple instructions
What is the primary advantage of RISC architectures in terms of performance?
Reduced clock cycles per instruction
How do CISC architectures typically handle complex instructions?
Through microprogrammed control units
What is the primary purpose of the register windows in a CISC architecture?
To enable fast data access and parameter passing
What is the main limitation of Flynn's Taxonomy of Computer Architecture?
It is limited in its ability to classify modern architectures
What is the main characteristic of the control unit in a RISC architecture?
It is hardwired for speed and has a simple instruction set
What is the primary advantage of pipelining and speculative execution in RISC architectures?
It reduces the number of clock cycles per instruction
What is the primary difference between RISC and CISC architectures in terms of memory access?
RISC architectures use load/store instructions only
What is the main characteristic of the instruction length in a RISC architecture?
Fixed length instructions
What is the primary characteristic of the number of registers in a RISC architecture?
Many registers, used for fast data access and parameter passing
What is the primary limitation of bus-based networks?
Performance bottleneck
What is a characteristic of tightly-coupled systems?
Shared memory
What is the primary distinction between SIMD and MISD architectures?
Number of data streams
What is the purpose of cache update protocols?
Maintaining cache coherence
What is a characteristic of Uniform Memory Access (UMA) systems?
All memory accesses take the same amount of time
What characterizes MIMD systems with shared memory?
Multiple processors sharing the same memory and communicating through memory access
What is a primary concern in cloud computing?
Security
What is the primary advantage of parallel processing?
Improved system throughput and fault tolerance
What is a characteristic of Non-uniform Memory Access (NUMA) systems?
Processors can access their own memory much faster than remote memory
What is superscalar design primarily characterized by?
Multiple execution units and instruction-level parallelism
What is the primary advantage of crossbar networks?
Non-blocking
What is the primary distinction between VLIW and superscalar architectures?
Compiler's ability to pack instructions
What is a characteristic of omega networks?
Blocking with less contention than bus-based networks
What is the primary advantage of VLIW architectures?
Reduced compiler conservatism and improved code visibility
What is the primary characteristic of vector processors?
Operate on entire vectors or matrices at once
What is a characteristic of distributed systems?
Loosely-coupled processing units
What is the purpose of Network of Workstations (NOWs)?
To enable resource transparency on remote machines
What is the primary consideration in MIMD system communication?
Topology and routing strategy
What type of interconnection networks are used in loosely-coupled distributed systems or tightly-coupled processor-to-memory configurations?
Multistage interconnection networks
What is the primary consideration in achieving linear speedup with additional processors in parallel processing?
Careful consideration of inherently sequential algorithm portions
Study Notes
Instruction Set Architectures (ISA)
- RISC (Reduced Instruction Set Computing) ISA: few, simple instructions, fixed-length, load/store instructions, hardwired control unit, pipelining, and speculative execution enabled
- CISC (Complex Instruction Set Computing) ISA: many, complex instructions, variable length, microprogrammed control unit, and less pipelining
Flynn's Taxonomy of Computer Architecture
- Classifies architectures based on the number of instruction streams and data streams they can handle simultaneously
- Categories:
- SISD (Single Instruction Stream, Single Data Stream)
- SIMD (Single Instruction Stream, Multiple Data Streams)
- MIMD (Multiple Instruction Streams, Multiple Data Streams)
- MISD (Multiple Instruction Streams, Single Data Stream)
- SPMD (Single Program, Multiple Data)
MIMD Architectures
- Categorized based on memory sharing and communication methods
- Shared Memory (SMP) vs. Distributed Memory (MPP) architectures
- Examples:
- NOW (Network of Workstations)
- COW (Collection of Workstations)
- DCPC (Dedicated Cluster Parallel Computer)
- P OPC (Pile of PCs)
Parallel Processing
- Increases system throughput and improves fault tolerance
- Limited by inherently sequential portions of algorithms
- Techniques:
- Pipelining: divides fetch-decode-execute cycle into stages
- Super-pipelining: uses pipeline stages requiring less than half a clock cycle
- Superscalar architectures: include multiple execution units, instruction fetch unit, and decoding unit
- VLIW (Very Long Instruction Word) architectures: pack independent instructions into one long instruction
Vector Processors and Interconnection Networks
- Operate on entire vectors or matrices at once
- Categorized by operand access: register-register, memory-memory
- Interconnection networks:
- Topology: significantly impacts message passing overhead
- Routing strategy and switching technique
- Examples: bus-based networks, crossbar networks, omega networks
Shared Memory Multiprocessors
- Tightly-coupled systems that share memory
- Processors do not necessarily share the same physical memory block
- Cache update protocols: write-through with update, write-through with invalidate, write-back
This quiz covers the differences between RISC (Reduced Instruction Set Computing) and CISC (Complex Instruction Set Computing) architectures, including memory access, instruction sets, and performance. Topics include explicit load/store instructions, variable length instructions, and pipelining and speculative execution.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free