Episode 1.6
INSIDE THE MACHINE
Class 11 (10+1) • Unit 1: Fundamentals of Computer
💡 Direct Stream: You can also watch directly on YouTube or via local server:
📺 Watch on YouTube.com
Episode code: 1.6
✍️ Episode 1.6 Topic Notes
📌 1: Welcome & Overview
- Topic: Inside the Machine — Memory Hierarchy & Storage Subsystems
- The Core Dilemma: CPUs compute in fractions of a nanosecond, but mass storage devices take milliseconds to retrieve data. How does computer architecture bridge this 100,000x latency divide?
- HP Board Exam Relevance:
- The Memory Hierarchy Pyramid: Speed vs. Capacity vs. Cost per bit (3-5 Marks).
📌 2: The Memory Hierarchy Pyramid
- The Fundamental Law: Fast memory is expensive and small; cheap memory is large but slow.
- The Multi-Tiered Pyramid (Top to Bottom):
📌 3: Primary Storage: SRAM vs. DRAM Architecture
- 1. Static RAM (SRAM):
- Storage Cell: 4 to 6 interconnected transistors configured as a bistable flip-flop circuit.
- Operation: Holds binary states (
0or1) continuously as long as power is applied. No refresh cycles required! - Performance: Extremely fast access time (1 – 5 ns); low latency; complex chip layout; high cost per bit.
📌 4: Read-Only Memory (ROM) and Firmware Evolution
- Non-Volatile Boot Storage: ROM retains critical bootstrap code (BIOS/UEFI, POST diagnostics) even when power is severed.
- 1. Mask ROM: Hardwired during silicon photolithography at the factory; permanent and unalterable.
- 2. PROM (Programmable ROM): Ships blank with microscopic internal fuse links; programmed once by blowing fuses permanently using a specialized high-voltage PROM burner.
- 3. EPROM (Erasable Programmable ROM):
📌 5: Cache Memory & The Principle of Locality
- Why Cache Exists: Mitigates the "von Neumann Memory Wall" — the severe processing bottleneck caused by high-speed CPU cores waiting for data transfers from slower external DRAM.
- The Principle of Locality of Reference:
- Temporal Locality: If a memory location is accessed now, the same location will likely be accessed again shortly (e.g., variables inside a programming
forloop). - Spatial Locality: If a memory location is accessed, nearby contiguous addresses will likely be accessed soon (e.g., sequential array elements or consecutive code instructions).
📌 6: Multi-Level Cache Architecture (L1, L2, L3)
- Level 1 (L1) Cache:
- Split into separate L1 Instruction Cache (L1i) and L1 Data Cache (L1d) to prevent bus contention.
- Integrated directly inside each CPU core die. Capacity: 32 KB – 64 KB per core. Latency: ~1 ns (1 – 4 clock cycles).
- Level 2 (L2) Cache: