Course

CS-471: Advanced multiprocessor architecture

Lectures in this course (23)

Covers Advanced Multiprocessor Architecture, discussing course logistics, components, grading, and trends in modern computing systems.

Project Presentation: Semester Projects and Topics

Covers semester projects, project logistics, successful project examples, proposal writing, and various project options related to systems, threading, serverless programming, virtual memory, cloud benchmarking, and AI accelerators.

Performance Metrics: Evaluation and Tools

Explores performance metrics, Amdahl's Law implications, means evaluation, challenges in multiprocessor systems, and commercial server software.

CMP Caches I

Explores CMP cache basics, transactional memory, in-flight transactions, cache overflow mechanism, and victim replication.

Parallel Programming I

Covers the basics of parallel programming, including concurrency, forms of parallelism, synchronization, and programming models like PThreads and OpenMP.

Reactive NUCA: Cache Block Placement & Replication

Explores cache block placement, replication policies, and management strategies in Reactive NUCA for multi-core caches.

Data-Parallel Programming: Vector & SIMD Processors

Explores data-parallel programming with vector processors and SIMD, and introduces MapReduce, Pregel, and TensorFlow.

Interconnects: Networks-on-Chip

Covers cache partitioning, Vantage scheme, interconnects, and Networks-on-Chip in many-core systems.

Distributed Memory

Covers distributed memory systems, cache organization, multi-socket architectures, memory coherence, and on-chip interconnect design.

Basic Cache Coherence

Introduces the basics of cache coherence, covering cache organization, coherence protocols, and challenges in multi-level cache hierarchies.

Advanced Cache Coherence

Explores advanced cache coherence techniques, emphasizing power-aware servers and optimization of coherence protocols.

Hardware Specialization

Explores hardware specialization, ASIC advantages over processors, and strategies for achieving ASIC-like efficiencies in chip design.

GPUs: Multithreading and Architecture

Explores GPUs' architecture, multithreading, and their role in machine learning, discussing limitations and future trends.

Memory Consistency I

Explores memory ordering in multiprocessor systems, discussing load/store semantics, total ordering, and different consistency models.

Memory Consistency II

Explores speculative memory consistency, challenges, solutions, performance overhead, and the impact of dynamic fence enforcement on achieving high performance.

Machine Learning Accelerators: Types and Efficiency

Explores machine learning workloads, DNN layers, systolic arrays, and the efficiency of specialized accelerators like TPUs.

Advanced Synchronization & Transactional Memory

Covers advanced synchronization techniques, atomic operations, lock implementations, and hardware support for queue locks.

Deep Learning Accelerators: Optimization Strategies

Explores optimization strategies for deep learning accelerators, emphasizing data movement reduction through batching, dataflow optimizations, and compression.

Server Processors: Efficiency and Design Optimization

Explores server processor inefficiencies, hardware integration trends, and design optimization criteria for efficient servers.

Emerging Memory I

Explores the landscape of big data, memory importance in online services, challenges faced by memory systems, emerging DRAM technologies, and storage-class memory.