Slides-Cap9-Part1-Computer Storage Devices

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

Where is non-volatile memory typically connected to?

CPU register

I/O buses (correct)

Cache memory

Main memory bus

What is the primary purpose of magnetic disks?

Cache memory

Non-volatile storage (correct)

Volatile storage

Random access memory

How do magnetic disks access information?

Using a CPU register to store data

Using a single motor to rotate all disks

Using a single read/write head for all surfaces

Using a separate read/write head for each surface (correct)

What is the characteristic of magnetic disks in terms of size, cost, and speed?

Big, cheap, and slow Signup and view all the answers

What is the addressing scheme used in magnetic disks?

Cylinder-head-sector addressing Signup and view all the answers

What was the use of floppy disks in the past?

As a device for physical data transport Signup and view all the answers

What is the smallest unit that can be addressed on a magnetic disk?

Sector Signup and view all the answers

What is the purpose of logical block addressing (LBA)?

To replace CHS addressing and enable larger storage capacities Signup and view all the answers

What is the term for all the concentric tracks under the read/write head at a given point on all surfaces?

Cylinder Signup and view all the answers

Why do modern disks have tracks with different numbers of sectors?

To get disks with bigger storage capacity Signup and view all the answers

What is the term for the time taken to position the arm over the proper track?

Seek time Signup and view all the answers

What is the typical range of tracks per surface on a magnetic disk?

5k ~ 30k tracks per surface Signup and view all the answers

What is the purpose of the read/write process step that waits for the desired sector to rotate under the read/write head?

To wait for the desired sector to rotate under the read/write head Signup and view all the answers

Why do inner tracks have less sectors than outer tracks?

Because they have a lower density Signup and view all the answers

What is the main difference between NOR and NAND flash memories?

NOR is randomly addressable, while NAND can only be read in blocks Signup and view all the answers

What is the process required before writing new data to flash memory?

Erase the memory block Signup and view all the answers

What is the characteristic of RAID 1?

Redundant and able to recover from one failure Signup and view all the answers

What is the purpose of the parity block in a RAID 5 array?

To enable recovery from a single drive failure Signup and view all the answers

What is the approximate read access time for a 256-byte transfer from flash memory?

6.5 µs Signup and view all the answers

What is the main advantage of NAND flash memories over NOR flash memories?

Higher storage density Signup and view all the answers

How many disks are required in a RAID 6 array to protect p disks?

p + 1 disks Signup and view all the answers

What was the cost of flash memory per gigabyte in 2010?

$2/GB Signup and view all the answers

What is the function of the Boolean XOR operation in a RAID 5 array?

To compute the parity of the data Signup and view all the answers

What is the limitation of flash memory in terms of writings?

1M recordings Signup and view all the answers

What is the characteristic of RAID 0?

Striped/interleaved volumes Signup and view all the answers

What is the purpose of the diagonals in a RAID 6 array?

To enable recovery from two drive failures Signup and view all the answers

What is the main reason for the decline of floppy disks and hard drives in mobile systems?

Solid state disks (SSD) Signup and view all the answers

What is the difference between RAID 5 and RAID 6?

RAID 6 provides additional fault tolerance Signup and view all the answers

How can the life cycle of flash memory be expanded?

Through the uniform distribution of writes through blocks Signup and view all the answers

What is the characteristic of RAID 2?

Applying memory-style error-correcting codes - ECC to disks Signup and view all the answers

What is the transfer time in magnetic disks?

The time it takes to transfer a block of bits under the r/w head Signup and view all the answers

What is the formula to calculate the average rotational latency (ARL)?

ARL = 0.5 × RotationPeriod Signup and view all the answers

What is the main difference between reliability and availability in RAID systems?

Reliability refers to the probability of failure, while availability refers to the probability of unavailability Signup and view all the answers

What is the purpose of mirroring in RAID systems?

To increase availability by copying data to identical disks Signup and view all the answers

What is the formula to calculate the disk access time (DAT)?

DAT = SeekTime + RotationalLatency + TransferTime + ControllerTime + QueuingDelay Signup and view all the answers

What is the primary advantage of using RAID systems?

Increased availability through the addition of redundant disks Signup and view all the answers

What is the definition of seek time in magnetic disks?

The time it takes for the r/w head to move to a specific sector Signup and view all the answers

What is the purpose of stripping in RAID systems?

To increase performance by spreading sequential data across multiple disks Signup and view all the answers

What is the trend in the cost of magnetic disks per GB over the years?

The cost has decreased from $100,000/GB to less than $0.5/GB Signup and view all the answers

What is the main characteristic of RAID 0?

It is a non-redundant array of disks that provides high performance Signup and view all the answers

Study Notes

Storage Overview

Non-volatile memory can be viewed as part of the memory hierarchy system or part of the I/O system.
It is connected to the I/O buses and not to the main memory bus.

Magnetic Disks

Purpose: non-volatile storage, big, cheap, and slow.
Lowest level in the memory hierarchy system.
Based on a rotating disk covered with a magnetic surface.
Use a read/write head per surface to access information.
Disks may have more than one platter.
Cylinder-head-sector addressing is used to access data.
A sector is the smallest unit that can be addressed.
Tracks and sectors numbers: 5k ∼ 30k tracks per surface, 100 ∼ 500 sectors per track.
Logical block addressing (LBA) is used instead of CHS.
Cylinder: all the concentric tracks under the r/w head at a given point on all surfaces.

Magnetic Disks Performance

Seek time: between 5 to 12 ms.
Rotational latency: 3,600 to 15,000 RPM, average rotational latency (ARL) is 8 ms to 2 ms.
Transfer time depends on transfer size per sector, rotation speed, recording density, and disk diameter.
Typical transfer rate: 3 to 65 MiB/s.

Magnetic Disks Evolution

Increase in the number of bits per square inch, i.e., density.
Steep price reduction: US$ 100,000/GB (1984) to less than US$ 0.5/GB (2012).
Considerable increase in RPM, from 3600 RPM (’80s) to close to 15000 RPM (2000’s).

RAID Systems

Disks differ from other levels of memory hierarchy because they are non-volatile.
They are also the lowest level, i.e., there is no other level to fetch on in the computer if the data is not on the disk.
Redundant array of independent disks (RAID) is used to improve availability and reliability.
RAID systems have multiple simultaneous accesses, data is spread into multiple disks, and stripping and mirroring are used to increase performance and availability.
Main characteristics: latency is not necessarily reduced, availability is enhanced through the addition of redundant disks, and lost information can be rebuilt through redundant information.
Reliability vs availability: reliability is less, i.e., more disks, greater fail probability, while availability is greater, i.e., failures do not necessarily lead to unavailability.

RAID Systems Levels

RAID 0: not redundant, but more efficient, striped/interleaved volumes.
RAID 1: redundant and able to recover from one failure, mirror/copy volume.
RAID 2: applies memory-style error-correcting codes (ECC) to disks, no commercial use.
RAID 3: bit-interleaved parity, one parity/check disk for multiple data disks, able to recover from one failure.
RAID 4: block-interleaved parity, one check disk for multiple data disks, able to recover from one failure.
RAID 5: distributed block-interleaved parity, able to recover from one failure.
RAID 6: RAID 5 extension, another parity block, able to recover from double faults.

Flash Memory

Technology similar to traditional EEPROM, higher memory capacity per chip.
Low-power consumption.
Read access time slower than DRAM, but much faster than disks.
Writing requires “deletion” of data, erase-before-write.
NOR- and NAND-based flash memories.
NAND flash memories have higher storage density, but can only be read in blocks.
NAND flash is much cheaper per gigabyte and much more common than NOR flash.
Price evolution: $2/GB (2010) to $0.3/GB (2016).
Wear-out of flash wrt writings, limited to 100K and 1M recordings.
Life cycle can be expanded through the uniform distribution of writes through blocks.
Solid state disks (SSD) extinguished floppy disks and hard drives in mobile systems.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

Learn about non-volatile memory and magnetic disks, including their purpose, characteristics, and how they work in the memory hierarchy system.