| Course Code: CS5289 Course Name: Advanced Memory Systems Prerequisites: Computer Architecture and Organization (CS2201 / CS223 or equivalent) Syllabus: Modern computing systems are often limited by memory performance, energy consumption, and data movement. Emerging technologies such as 3D-stacked memories and non-volatile memories are changing the design of memory systems. This course introduces modern memory architectures including DRAM systems, 3D-stacked memories, and emerging non-volatile memories. It also discusses performance, energy, and thermal challenges, and system-level techniques for efficient memory usage in workloads such as machine learning and data-intensive applications. Content: Memory System Fundamentals: Memory hierarchy, locality, latency, bandwidth, energy, memory wall; DRAM Architecture and Operation: DRAM organization, banks, rows, timing parameters, refresh mechanisms, memory controllers; Memory System Optimization: Row buffer management, memory scheduling policies, memory-level parallelism, address mapping techniques; Advanced DRAM Systems: Multi-channel memories, high-bandwidth memory systems, memory scalability issues; 3D-Stacked Memories: TSV-based integration, architectures such as HBM and HMC, architectural benefits and design challenges; Thermal Challenges in 3D Systems: Power density, thermal hotspots, thermal modeling, dynamic thermal management techniques; Emerging Non-Volatile Memories: PCM, STT-MRAM, ReRAM, device characteristics, latency, endurance and retention trade-offs; Hybrid Memory Systems: DRAM?NVM integration, data placement policies, write reduction techniques, endurance management; Memory Systems for Modern Workloads: Memory challenges in machine learning and data-intensive applications; Research Trends: Case studies from recent architecture research on 3D memory systems and NVM-based computing. Texts: 1. Bruce Jacob, David Wang & Spencer Ng (2010). Memory Systems: Cache, DRAM, Disk. Morgan Kaufmann 2. Rasit O. Topaloglu (Ed.) (2015). More than Moore Technologies for Next Generation Computer Design. Springer New York 3. John L. Hennessy & David A. Patterson (2011). Computer Architecture: A Quantitative Approach. Elsevier |