Computer Science: Disk Storage and RAM

Computer Organization and Architecture

Unit-III: Memory and I/O

• The time taken to read a block of data from a disk into memory is composed of:
  • Seek time
  • Rotational latency
  • Transfer time
• Rotational latency refers to the time it takes for the platter to make a full rotation.
• RAM memory is not suitable for permanent storage because it is volatile.
• A Flip Flop is a circuit used to store one bit of data.
• Access time is the average time required to reach a storage location in memory and obtain its contents.
• The idea of cache memory is based on the property of locality of reference.
• Cache memory acts between the CPU and RAM.
• Write Through technique is used in cache memory for updating data.
• Dynamic RAM is used as main memory in a computer system because it has higher speed.
• Virtual memory consists of Magnetic memory.
• Cache memory works on the principle of locality of reference.
• Memory unit accessed by content is called Associative Memory.

Unit-IV: Microprocessor

• The advantage of memory mapped I/O over I/O mapped I/O is that it is faster.
• In EU (Execution Unit) of the 8086 microprocessor, IP (Instruction Pointer) is not present.
• ALE (Address Latch Enable) is used to demultiplex the Address bus of 8086.
• The signal used to control the direction of data flow of bus transceivers is DT/R (Data Transmit/Receive).
• LOCK is a signal that belongs to maximum mode only.
• The invalid pair of registers is CS:SP.
• The Trap Flag is used for single stepping.
• The three basic sections of a microprocessor unit are:
  • Control and timing
  • Register
  • Arithmetic/logic unit (ALU)
• The data bus is a bidirectional bus.
• Direction flag is used with String instructions.
• Microprocessor speed depends on the clock.
• A shift register is a register capable of shifting its binary information either to the right or the left.
• Maskable interrupts are interrupts that can be turned off by the programmer.
• In a DMA write operation, the data is transferred from I/O to memory.

Unit V: Pipelining and Parallel Processing

• Optimizing compilers have been developed specifically for pipelined systems.
• The pipelining process is also called as Assembly line operation.
• The fetch and execution cycles are interleaved with the help of the clock.
• Each stage in pipelining should be completed within 1 cycle.
• To increase the speed of memory access in pipelining, we make use of Cache.
• Pipelining is a parallel operation technique.
• In a DMA write operation, the data is transferred from I/O to memory.
• Multiplying two 5x5 matrices using a vector processor requires 5 product terms in each inner product.
• SIMD array processor is a type of array processor.
• The full form of PE in SIMD array processor is Processing Element.
• The full form of SIMD is Single Instruction stream and Multiple Data stream.
• An SIMD array processor is a processor that has single instruction multiple data organization.
• FLOPS (Floating Point Operations Per Second) is a measure used to evaluate computers in their ability to perform a given number of floating-point operations per second.
• There are 2 types of multiprocessors.
• A disadvantage of Multiprocessor Systems is that all of the above (expensive, require a much larger pool of memory, and more complex operating system).
• RISC (Reduced Instruction Set Computing) processor requires more number of registers.
• Multiplier pipeline and Adder pipeline are used to calculate an inner product in a vector processor.
• The term gigaflops is used to denote billion flops.
• Stack-organized architecture is used in Zero Address Instruction.
• Asymmetric Multiprocessor system contains a master-slave relationship.
• A multiprocessor operating system must take care of Unauthorized data access and data protection.