Parallel and Distributed Computing - AP Computer Science Principles

Computing using multiple networked computers to complete tasks. Spreads computation across separate machines connected via network.

A single device in a distributed system. Individual computer or device participating in the distributed network.

Even distribution of work across processors to optimize performance. Prevents any single processor from being overwhelmed while others idle.

Ability to continue operation despite failures. System maintains functionality when individual components fail.

Amount of work performed by a system in a given time period. Measures data processed or tasks completed per unit time.

Network of computers working together to perform large tasks. Geographically distributed computers share computational workload efficiently.

Time delay in data processing or communication. Measures time between request initiation and response completion.

Extremely fast computer for complex calculations. Specialized machine designed for high-performance scientific computations.

Software layer that enables communication between components. Abstracts hardware differences to simplify distributed application development.

Increased complexity in programming. Coordinating multiple processors requires sophisticated programming techniques.

Software that creates and manages virtual machines. Enables multiple operating systems to run on single physical machine.

Cloud computing services. Remote servers provide scalable computing resources on demand.

$S = \frac{T_s}{T_p}$, where $T_s$ is serial time, $T_p$ is parallel time. Measures performance improvement from using multiple processors.

Multiple computations are executed during overlapping time periods. Tasks progress simultaneously but may not execute at identical moments.

A point of congestion that limits performance. Slowest component determines overall system performance speed.

Flawed execution outcome due to unsynchronized data access. Multiple processes competing for resources produce unpredictable results.

Communication method between processes via messages. Processes exchange information without sharing memory directly.

Observation that transistor counts double approximately every two years. Predicts exponential growth in computing power over time.

Use of different types of processors for computing tasks. Combines CPUs, GPUs, and specialized processors for optimal performance.

Coordination of events to operate in unison. Ensures processes execute in correct order for consistent results.

Database in which data is stored across multiple locations. Information spans multiple sites for redundancy and accessibility.

A single computing component with two or more independent cores. Multiple processing units on single chip enable parallel execution.

Distributing tasks across multiple systems to balance workload. Prevents system overload by distributing work among available resources.

To allocate resources to tasks in a computing environment. Determines execution order and resource allocation for optimal performance.

Computing environment where all processors are of the same type. All processors share identical architecture and instruction sets.

Extremely fast computer for complex calculations. Specialized machine designed for high-performance scientific computations.

Multiple computations are executed during overlapping time periods. Tasks progress simultaneously but may not execute at identical moments.

Distributing tasks across multiple systems to balance workload. Prevents system overload by distributing work among available resources.

Ability to handle increasing workloads by adding resources. System grows efficiently as demand increases without performance loss.