Chapter 1: Computer Abstractions and Technology

Vocabulary flashcards covering key terms and concepts from Chapter 1: Computer Abstractions and Technology.

RISC-V

An open Instruction Set Architecture (ISA) with 32 general-purpose registers and 32 floating-point registers; byte-addressed memory; supports base+offset addressing and immediate/register addressing with fixed-length encoding.

ISA

Instruction Set Architecture; the interface between hardware and software, defining registers, addressing modes, operations, and encoding.

ABI

Application Binary Interface; conventions and formats governing binary interaction between program and system software/hardware.

Abstraction

A design principle that hides low-level details to simplify complexity and manage system design.

Datapath

The part of the CPU that performs operations on data.

Control

The part of the CPU that sequences the datapath, memory, and I/O operations.

Cache memory

Small, fast SRAM memory used to store data for quick access and reduce latency.

Volatile memory

Main memory that loses its contents when power is removed.

Non-volatile memory

Secondary storage that retains data without power, e.g., magnetic disks, flash memory, optical disks.

Byte-addressed memory

Memory where addresses refer to individual bytes; aligned accesses are faster.

Base+offset addressing

An addressing mode using a base register plus a fixed offset to form a memory address.

Load-store architecture

An architecture where operations mostly occur on registers, with memory access via load/store instructions.

Memory hierarchy

Organized layers of memory (registers, caches, main memory, secondary storage) to balance speed and cost.

PMD

Personal Mobile Device; battery-powered, internet-connected devices like smartphones and tablets.

WSC

Warehouse Scale Computers; large data-center systems used to run cloud-based services.

SaaS

Software as a Service; software delivered over the cloud, with parts running on PMDs and in the cloud.

DLP

Data-Level Parallelism; performing the same operation on multiple data elements simultaneously.

TLP

Task-Level Parallelism; parallelism across distinct tasks or threads.

ILP

Instruction-Level Parallelism; overlapping execution of multiple instructions by the hardware.

Flynn’s taxonomy

Classification of computer architectures by instruction and data streams: SISD, SIMD, MISD, MIMD.

SISD

Single Instruction Stream, Single Data Stream.

SIMD

Single Instruction Stream, Multiple Data Streams; used by vector processors and GPUs.

MISD

Multiple Instruction Streams, Single Data Stream; rarely implemented in commercial systems.

MIMD

Multiple Instruction Streams, Multiple Data Streams; can be tightly or loosely coupled.

Amdahl’s Law

The principle that overall speedup is limited by the serial portion of the system; improving the common case yields the most gain.

Seven Great Ideas

Key design principles: abstraction, fast common case, parallelism, pipelining, prediction, memory hierarchy, dependability via redundancy.

Locality

Principle of locality: reuse of data and instructions, enabling effective caching.

Benchmark

A standard set of programs used to measure and compare performance, including kernels, toy programs, and benchmark suites.

Execution time

Total time to complete a task, including processing, memory, I/O, and OS overhead.

CPU time

Time spent performing computations (user + system time), excluding I/O and idle time.

CPI

Average cycles per instruction; depends on the instruction mix and CPU hardware.