Articles, deep dives, and notes from the engineering community.
A study of multicore scalability, exploring Symmetric Multiprocessing (SMP), Non-Uniform Memory Access (NUMA), Amdahl's law, and on-chip mesh Networks-on-Chip.
By Md. Harun-Or-Roshid
A study of data-level parallelism, exploring Flynn's taxonomy, AVX-512 vector lane striping, GPU Streaming Multiprocessors, warps, and SIMT execution engines.
A masterclass in computer instruction set architectures, comparing x86 CISC with ARM, evaluating the Iron Law of CPU performance, and exploring modular open RISC-V.
A comprehensive analysis of virtual memory architecture, multi-level hierarchical page tables, hardware MMU address translation, and TLB HugePages acceleration.
A deep dive into high-speed memory systems, examining SRAM cache organizations, direct and set-associative mappings, write policies, and multicore MESI coherence.
A masterclass in CPU pipeline hazard mitigation, detailing structural Harvard splits, RAW data forwarding bypass units, load-use bubbles, and dynamic branch prediction.
A study of CPU instruction pipelining, exploring the classic 5-stage RISC datapath, inter-stage pipeline registers, and throughput speedup mathematics.
A comparative analysis of single-cycle and multi-cycle processor datapaths, critical path timing calculations, and hardwired vs microprogrammed control units.
An in-depth analysis of processor ALU circuit engineering, comparing ripple-carry adders with carry-lookahead logic, Booth multiplication, and IEEE 754 FPUs.
A deep architectural analysis of computer interconnection structures, examining data, address, and control buses, bus arbitration, and high-speed PCIe lanes.