Lecture

Advanced Multiprocessor Architecture

Related lectures (32)

GPU Advanced: Memory Management and Optimization

Explores advanced GPU memory management, CUDA programming, and data-parallel computing for optimizing performance.

Neuromorphic Computing: Challenges & Future

Compares digital and neuromorphic computing, highlighting challenges and future prospects in the field.

Covers the principles of synchronization in parallel computing, focusing on shared memory synchronization and different methods like locks and barriers.

VLIW Architectures: Compilers and Instruction Level Parallelism

Covers VLIW architectures and compilers, focusing on instruction level parallelism and optimization techniques.

GPU Memory Hierarchy and Optimization

Explores GPU memory hierarchy, optimization challenges, and algorithm efficiency.

Technological Evolution: CPUs & Storage

Explores the evolution of CPUs and storage technologies, RAM bit-flipping, Big Data applications, and technical challenges.

Implementation with Transistors

Covers the implementation of a processor with transistors and focuses on improving performance through various technologies and circuit designs.

Synchronization: Locks and Barriers

Explores synchronization principles using locks and barriers, emphasizing efficient hardware-supported implementations and coordination mechanisms like OpenMP.

Synchronization: Hardware Implementation

Explores hardware synchronization methods, including locks, barriers, and critical sections in parallel computing.

Challenge of even bigger models

Explores simulating large-scale neural network models and optimizing memory efficiency in neural simulations using NEURON and CoreNEURON.

Scaling Trends: Parallel Hardware Evolution

Explores the evolution of parallel hardware, from Moore's Law to the challenges of the 2000s, and the shift towards scale-out systems in data centers.

Transactional Memory: Two-phase Locking and Invisible Reads

Explores historical perspectives and mechanisms of transactional memory, emphasizing the importance and challenges of its implementation in modern computing systems.