Podcast
Questions and Answers
Which of the following types of RAM requires periodic charge refreshing to maintain data storage?
Which of the following types of RAM requires periodic charge refreshing to maintain data storage?
- DRAM (correct)
- ROM
- PROM
- SRAM
SRAM is typically favored for large memory requirements.
SRAM is typically favored for large memory requirements.
False (B)
What are the three main ways DDR SDRAM achieves higher data rates?
What are the three main ways DDR SDRAM achieves higher data rates?
DDR SDRAM achieves higher data rates by transferring data on both the rising and falling edge of the clock, utilizing a higher clock rate on the bus, and employing a buffering scheme.
A permanent physical defect in a memory cell is known as a ______ failure.
A permanent physical defect in a memory cell is known as a ______ failure.
Match the following memory types with their primary characteristics:
Match the following memory types with their primary characteristics:
Which of the following memory types is most suitable for high-volume production runs, offering flexibility and convenience?
Which of the following memory types is most suitable for high-volume production runs, offering flexibility and convenience?
Flash memory is a type of volatile memory.
Flash memory is a type of volatile memory.
Which type of memory is typically used in solid state drives (SSDs)?
Which type of memory is typically used in solid state drives (SSDs)?
Solid state drives (SSDs) are generally faster than traditional hard disk drives (HDDs).
Solid state drives (SSDs) are generally faster than traditional hard disk drives (HDDs).
What is a major drawback of SSDs in comparison to HDDs?
What is a major drawback of SSDs in comparison to HDDs?
The ______ is the interface that connects the SSD to the host system.
The ______ is the interface that connects the SSD to the host system.
Match the following memory types with their respective characteristics:
Match the following memory types with their respective characteristics:
Which of these is NOT a technique used to prolong the lifespan of flash memory in SSDs?
Which of these is NOT a technique used to prolong the lifespan of flash memory in SSDs?
A compact disk (CD) is an example of an external memory device.
A compact disk (CD) is an example of an external memory device.
Explain the purpose of a data buffer/cache in an SSD.
Explain the purpose of a data buffer/cache in an SSD.
The process of evenly distributing writes across different blocks of flash memory cells is known as _______.
The process of evenly distributing writes across different blocks of flash memory cells is known as _______.
Which of the following is considered a secondary storage device?
Which of the following is considered a secondary storage device?
Which RAID level provides the highest data availability, requiring three disks to fail before data loss occurs?
Which RAID level provides the highest data availability, requiring three disks to fail before data loss occurs?
In RAID 1, a read request can be serviced by either of the two disks containing the requested data.
In RAID 1, a read request can be serviced by either of the two disks containing the requested data.
What is the key difference between RAID levels 4 and 5 in terms of parity strip distribution?
What is the key difference between RAID levels 4 and 5 in terms of parity strip distribution?
RAID 3 uses a simple ______ bit to compute redundancy for the set of individual bits in the same position on all data disks.
RAID 3 uses a simple ______ bit to compute redundancy for the set of individual bits in the same position on all data disks.
Match the following RAID levels with their characteristics:
Match the following RAID levels with their characteristics:
Flashcards
Dynamic RAM (DRAM)
Dynamic RAM (DRAM)
A type of RAM that stores data as charge in capacitors and needs refreshing.
Static RAM (SRAM)
Static RAM (SRAM)
A type of RAM built with flip-flops that retains data as long as power is supplied.
Read Only Memory (ROM)
Read Only Memory (ROM)
Non-volatile memory that contains permanent data that cannot be changed.
Programmable ROM (PROM)
Programmable ROM (PROM)
Signup and view all the flashcards
Synchronous DRAM (SDRAM)
Synchronous DRAM (SDRAM)
Signup and view all the flashcards
Double Data Rate SDRAM (DDR SDRAM)
Double Data Rate SDRAM (DDR SDRAM)
Signup and view all the flashcards
Flash Memory
Flash Memory
Signup and view all the flashcards
SSD Cost per GB
SSD Cost per GB
Signup and view all the flashcards
HDD Cost per GB
HDD Cost per GB
Signup and view all the flashcards
Flash Memory Lifespan
Flash Memory Lifespan
Signup and view all the flashcards
SSD Performance Issues
SSD Performance Issues
Signup and view all the flashcards
Write Operations
Write Operations
Signup and view all the flashcards
Wear-Leveling Algorithms
Wear-Leveling Algorithms
Signup and view all the flashcards
Bad-Block Management
Bad-Block Management
Signup and view all the flashcards
Flash Memory Architecture
Flash Memory Architecture
Signup and view all the flashcards
RAM Buffer
RAM Buffer
Signup and view all the flashcards
Compact Disk
Compact Disk
Signup and view all the flashcards
RAID Level 1
RAID Level 1
Signup and view all the flashcards
RAID Level 2
RAID Level 2
Signup and view all the flashcards
RAID Level 3
RAID Level 3
Signup and view all the flashcards
RAID Level 4
RAID Level 4
Signup and view all the flashcards
Advantages of SSDs
Advantages of SSDs
Signup and view all the flashcards
Study Notes
Internal Memory
- RAM technology is divided into two types: Dynamic RAM (DRAM) and Static RAM (SRAM)
- DRAM is made with cells that store data as charge on capacitors
- The presence or absence of charge in a capacitor is interpreted as binary 1 or 0
- DRAM requires periodic charge refreshing to maintain data storage
- The term "dynamic" refers to the tendency of the stored charge to leak away, even with continuous power application
- SRAM uses the same logic elements as a processor
- Binary values are stored using traditional flip-flop logic gate configurations
- SRAM holds its data as long as power is supplied
SRAM versus DRAM
- Both SRAM and DRAM are volatile, meaning power must be continuously supplied to preserve bit values
- DRAM cells are simpler to build and smaller, resulting in more cells per unit area, making them less expensive
- DRAM requires supporting refresh circuitry
- DRAM is favored for large memory requirements and used for main memory
- SRAM is faster and used for cache memory (both on- and off-chip)
Read Only Memory (ROM)
- ROM contains a permanent pattern of data that cannot be changed or added to
- No power source is needed to maintain bit values
- Data or programs are permanently stored in main memory and never need loading from secondary storage
- Data is wired into the chip during fabrication
- Disadvantages include no room for error, and a relatively high fixed cost for data insertion
Programmable ROM (PROM)
- A less expensive alternative to ROM
- PROM is non-volatile and can be written into only once
- The writing process is electrical and can be performed by a supplier or customer after the original chip fabrication
- Special equipment is required for programming
- PROM provides flexibility and convenience, and is attractive for high-volume production runs
Read-Mostly Memory
- This section includes categories of memory that fall in between EPROM and EEPROM in both cost and functionality
- EPROM: Erasable Programmable Read-Only Memory. Erasure can be performed repeatedly. More expensive than PROM but has multiple update capabilities.
- EEPROM: Electrically Erasable Programmable Read-Only Memory. Can be written into at any time without erasing prior contents. More expensive than EPROM.
- Flash: Combines advantages of non-volatility with the flexibility of being updatable in place. Uses electrical erasing. Does not provide byte-level erasure. Uses only one transistor per bit, achieving high density.
Error Correction
- Hard Failures: Permanent physical defects in a memory cell. Cells may become stuck at 0 or 1, or switch erratically between 0s and 1s. Can be caused by environmental abuse, manufacturing defects, or wear.
- Soft Errors: Random, non-destructive events that alter memory cell contents. Cause includes power supply problems and alpha particles.
Advanced DRAM Organization
- One of the most critical system bottlenecks using high-performance processors is how the memory interface works with the processor.
- Traditional DRAM chips are constrained by both their internal architecture and their interface with the processor’s memory bus
- Improvements to the basic DRAM architecture have been explored and schemes that currently dominate the market are SDRAM and DDR-DRAM
Synchronous DRAM (SDRAM)
- SDRAM exchanges data with the processor synchronized with an external clock signal
- It runs at the full speed of the processor/memory bus without imposing wait states.
- The processor (or other master) initiates the instruction and address which the DRAM latches
- DRAM responds after a set number of clock cycles
- The master can safely perform other tasks while the SDRAM processes
Double Data Rate SDRAM (DDR SDRAM)
- Developed by JEDEC
- Companies make DDR chips which are widely used in desktop computers and servers.
- DDR achieves higher data rates via:
- Synchronization to both the rising and falling edges of the clock.
- Using a higher clock rate on the bus.
- Using a buffering scheme.
Flash Memory
- Used for both internal and external memory applications.
- Introduced in the mid-1980s.
- Intermediate between EPROM and EEPROM in cost and functionality
- Uses electrical erasing technology,
- Erasing is possible for blocks, not entire chips.
- Does not provide byte-level erasure, but achieves high density because of transistor use
- Microchip is organized to erase sections of cells ("flash").
Memory Hierarchy
- A pyramid shows a hierarchy of NVMMs (Non-volatile memory modules) from highest performance (SRAM), decreasing cost per bit, and increasing capacity/density to the lowest performance (hard disk).
- Includes SRAM, STT-RAM, DRAM, NVM flash, PCRAM, ReRAM, hard disk
External Memory
Magnetic Disk
- A disk is a circular platter of non-magnetic material coated with a magnetizable material
- Traditionally, substrates are aluminum or aluminum alloy.
- Glass substrates are also used to improve uniformity of the magnetic film and reduce read/write errors, stiffness for reduced dynamics, higher shock resistance
- The write mechanism relies on the magnetic field generated by electricity flowing through a coil to magnetize regions on the disk.
- An electric current changes direction to reverse magnetization on the medium
Magnetic Read and Write Mechanisms
- Data is recorded on or retrieved from the disk via a conducting coil called the head, often a read head and write head
- Electric pulses sent to the write head create magnetic patterns on the disk
- Different patterns correspond to positive and negative currents
- Magnetizing a small area on the disk reverses the current direction
Disk Data Layout
- Data is organized into sectors, tracks, cylinders, and a read-write head
- The heads are on an arm, which can move to specific locations.
- The platter rotates under the stationary arm/head during reading/writing
- The arrangement (figure) shows a representation of how data is stored on the disk (i.e., tracks/sectors).
Physical Characteristics of Disk Systems
- Head motion types: Fixed (one per track) or movable (one per surface).
- Disk portability: Non-removable (e.g., hard disk) or removable (e.g., floppy).
- Sides: Single or double-sided.
- Head mechanism: Contact (floppy), fixed gap, or aerodynamic gap (Winchester).
- Physical characteristics of disk types include platter types, head position, etc.
Disk Classification
- Winchester heads: Used in sealed assemblies, free from contaminants for close operation to the disk and high data density
- The mechanism uses an aerodynamic foil that moves on the platter's surface even when the disk is motionless.
- The air pressure of the spinning disk prevents the foil from touching the surface, thus allowing for near-surface operation and high data density.
- The head must generate or sense an electromagnetic field of sufficient magnitude to write/read properly.
- Narrower heads require closer operation to the platter surface, which leads to tracks and higher data density
Typical Hard Disk Drive Parameters
- Table shows specifications for different types of hard disk drives (including capacity, average seek time, spindle speed, average latency, maximum transfer rate, byte/sectors, tracks, cache size)
Timing of a Disk I/O Transfer
- Shows sequence of events in a disk transfer (wait for device, wait for channel, seek operation, rotational delay, data transfer)
Disk Performance Parameters
- Seek time: time required for the head to move to the correct track
- Rotational latency: time required for the desired sector to reach the head
- Access time: total time required for the read/write head to position at the correct track and sector, including seek time and rotational latency
- Transfer time: time for the data transfer
RAID
- RAID is an acronym for Redundant Arrays of Independent Disks.
- Redundant array of independent disks (RAID) consists of multiple hard drives, combined to perform as a single volume.
- Provides data redundancy and improved performance over a single hard drive
RAID Levels
- RAID Level 0: Data striping without redundancy (higher throughput). Data striping across multiple disks. Data lost if one drive fails.
- RAID Level 1: Data mirroring (redundancy). Creates a redundant copy of all data. High data availability as data recovery can be done from a backup copy if a drive fails.
- RAID Level 2: Error correction using Hamming code (highest performance). Uses a parallel access technique, where data is distributed across multiple disks
- RAID Level 3: Data striping with parity (moderate cost and availability). Employs parallel access, with data distributed across the drives, and parity data dispersed in a distributed manner. Only one dedicated parity drive is needed
- RAID Level 4: Data striping with parity on a block (moderate/reduced cost/availability). Data operations handled independently to satisfy I/O request in parallel
- RAID Level 5: Data striping with distributed parity (high/increased cost/availability). Parity strips distributed across all disks to avoid a bottleneck .
- RAID Level 6: Data striping with dual parity (higher/increased cost/availability). Two different parity calculations are performed and stored in separate blocks on different disks. Very high data availability
SSD Compared to HDD
- SSDs have advantages over HDDs, including high-performance input/output operations per second (IOPS), durability, and longer lifespan. Also have lower power consumption, quieter and cooler running, and lower access times.
NAND Flash Drives
- Comparison table for NAND flash drives and HDDs, including file copy/write speed, power draw, storage capacity, and cost.
- Differences in speed and power consumption with HDDs. .
Optical Disk Products
- Includes CD-ROM, CD-R, CD-RW, DVD, DVD-R, DVD-RW, and Blu-Ray
- CD-ROM is non-erasable, and good for distribution of large data amounts
- CD-R can be written to only once
- CD-RW is rewritable
- DVD products have larger capacities than CD products
- Blu-Ray uses a shorter wavelength laser, allowing for higher storage density
Compact Disk Read-Only Memory (CD-ROM)
- CD-ROM technology similar to audio CDs
- Different in that an error correction device ensures that transfers are properly transferred
CD-R, CD-RW
- CD-Recordable (CD-R) - One time writing
- CD-Rewritable (CD-RW) – Multi-time writing
Magnetic Tape
- Tape systems use similar read/write techniques as disk systems
- Medium is flexible polyester coated with a magnetizable material
- Data is stored on parallel tracks
- Data stored in blocks called "physical records"
- Records separated by gaps("inter-record gaps").
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.