Ism Module 4: Intelligent Storage Systems (Iss)

Front 1

Name four types of Intelligent Storage Systems

Back 1

1. Block-based storage system

2. File-based storage system

3. Object-based storage system

4. Unified storage system

Front 2

What is an Intelligent Storage System?

Back 2

A feature-rich RAID array that provides highly optimized I/O processing capabilities.

Front 3

Features of an Intelligent Storage System

Back 3

- Supports combination of HDD and SSD
- Service massive amount of IOPS

- Scale-out architecture

- Deduplication, compression, and encryption

- Automated storage tiering

- Virtual storage provisioning

- Multi-tenancy

-Supports APIs to integrate with SDDC and cloud

-Data protection

Front 4

Large amount of memory

Back 4

Cache

Front 5

Two key components of an Intelligent Storage System

Back 5

Controller

Storage

Front 6

Types of Controller components within an ISS

Back 6

Block-based

File-based

Object-based

Unified

Front 7

Types of Storage components within an ISS

Back 7

HDDs

SSDs

Combination of both

Front 8

Four Components that make up a Storage Array (ISS) (Jeffy's Definition)

Back 8

F.E. Ports

Cache

B.E. Ports

Disks

Front 9

What is a controller?

Back 9

A controller is a compute system that runs a purpose-built OS that is responsible for performing several key functions for the storage system.

Front 10

Components of a Hard Disk Drive

Back 10

Platter

Spindle

Read/Write Head

HDA--> Head Disk Assembly

Actuator Arm Assembly

Drive Controller Board

Front 11

Characteristics of a Platter

Back 11

- Data recorded on these platters is in binary code.

- Data is encoded by polarizing the magnetic area or domains of the disk.

- Data can be written to or read from both surfaces.

- The number of platters and the storage capacity of each platter determine the total capacity of the drive.

Front 12

Head Disk Assembly (HDA)

Back 12

A set of rotating platters is sealed in this case.

Front 13

What is a Spindle?

Back 13

- A spindle connects all the platters and is connected to a motor.

Front 14

Function of the Read/Write Head

Back 14

Read and write data from or to the platters.

Drives have two R/W heads per platter, one for each surface of the paltter.

Front 15

Head Flying Height

Back 15

A microscopic air gap that is maintained between the R/W heads and the platters.

Front 16

What is a Head Crash?

Back 16

- The magnetic coating on the platter is scratched and may cause damage to the R/W/ head.

- A head crash generally results in data loss.

Front 17

Actuator Arm Assembly

Back 17

R/W heads are mounted on the actuator arm assembly, which positions the R/W/ heads at the location on the platter where the data needs to be written or read.

Front 18

What does the Drive Controller Board consist of?

Back 18

- Controller Board consists of microprocessor, internal memory, circuitry, and firmware.

- The controller is a printed circuit board, mounted at the bottom of the disk drive.

Front 19

Tracks per Inch (TPI)

Back 19

The number of tracks per inch (TPI) on the platter (on track density) measure how tightly the tracks are packed on a platter.

Front 20

Track is divided into

Back 20

Sectors

A sector is the smallest, individually addressable united of storage.

Front 21

How large is a sector typically?

Back 21

512 bytes of user Data

Front 22

A customer can never asked you the size of this

Back 22

Track- the concentric ring on the platter around the spindle.

Front 23

Cylinder, Head, and Sector (CHS) Number

Back 23

Refers to specific physical address location on the disk.

EX. Student NAME

Front 24

Logical Block Address (LBA)

Back 24

Simplified the addressing by using a linear address to access physical blocks of data.

EX. Student 1

Front 25

Calculation of Blocks

Back 25

Sectors per track * heads * cylinders

Block Range 0- ^ that number

Front 26

Time taken by a disk to complete an I/O request

Back 26

Seek Time

Rotation latency

Data Transfer Rate

Front 27

Disk Service Time Formula

Back 27

Seek Time + Rotational Latency + Data Transfer Time

Front 28

What is Seek Time?

Back 28

Describe the time taken to position the R/W heads across the platter with a radial movement (moving along the radius of the platter)

(also called access time)

Front 29

Full Stroke Seek Time

Back 29

The Time taken by the R/W/ head to move across the entire width of the disk, from the innermost track to the outermost track.

(Measured in ms)

Front 30

Average Seek Time

Back 30

The average time taken by the
R/W head to move from one random track to another, normally listed as the time for one-third of a full stroke.

(Measured in ms)

Front 31

Track-to-track Seek Time

Back 31

It is the time taken by the R/W head to move between adjacent tracks.

(Measured in ms)

Front 32

Short-Stroking

Back 32

Results in lower usable capacity than the actual capacity of the drive.

EX. 500 GB disk set up to use the first 40%, now treated as a 200 GB Drive.

Front 33

Average Rotational Latency Formula

Back 33

(One-half of the time taken for a full rotation)

(1/2*1000)/ (X/60)

Front 34

What is Rotational Latency?

Back 34

The time taken by the platter to rotate and position the data under the R/W Head.

Front 35

Data Transfer Rate

Back 35

Average amount of data per unit time that the drive can deliver to the HBA.

Front 36

Internal transfer rate

Back 36

The speed at which data moves from a platter's service to the internal buffer (cache) of the disk

Front 37

External Transfer Rate

Back 37

The rate at which data can move through the interface to the HBA.

Front 38

Average Response Time Formula

Back 38

Service Time/ (1-Utilization)

Front 39

What is a Queue?

Back 39

The location where an I/P request waits before it is processed by the I/O controller and disk I/O controller processes I/Os waiting in the queue one by one.

Front 40

Disks required to meet an application's capacity need (DC)

FORMULA

Back 40

Dc= Total capacity required/ Capacity of a single disk.

Front 41

Disks required to meet applications performance need (DP)

FORMULA

Back 41

Dp= IOPS generated by an application at peak workload/ IPOS serviced by a single disk.

Front 42

IOPS serviced by a disk (S) depends upon disk service time (Ts)

FORMULA

Back 42

Ts= Seek Time + (0.5/ (Disk rpm/60)) + (Data Block size/ Data tranfers rate)

Front 43

TS is time taken for an I/O to complete, therefore IPOS services by a disk (S) is equal to (1/TS)

FORMULA

Back 43

S= 0.7 * (1/Ts)

Front 44

Disk required for application

Back 44

Max (DC, DP)

Front 45

Components of a Solid State Drive (SSD)

Back 45

I/O Interfact

Controller ( RAM Cache, Drive Controller, Non Volatile Memory)

Mass Storage (Disk)

Front 46

What is a SSD?

Back 46

Storage device that contain non-volatile flash memory

Front 47

What is non-volatile RAM used for?

Back 47

The non-volatile RAM (NVRAM) is used to store the SSD's operation softwareand data.

Front 48

Write coalescing

Back 48

A technique employed within RAM, the process of grouping write I/Os and writing them in a single internal operation versus many smaller-sized write operations.

Front 49

Flash memory

Back 49

Retain their contents when powered off.

The mass storage of an array of non-volatile flash memory chips.

Front 50

Single Level Cell (SLC)

Back 50

SLC- Type flash typically used in enterprise-rated SSSs for its increased memory speed and longevity.

Front 51

Multi-Level Cell (MLC)

Back 51

Slower than SLC, but the advantage of a greater capacity per chip.

Front 52

Solid State Drive stores bits

Back 52

Eight bits make up a byte

Front 53

Page

Back 53

Physical data object

Is the smallest object that can be read or written on a solid state drive.

Pages do NOT have a standard capacity

Front 54

Solid State Drive Block

Back 54

Made up of pages.

32, 64, 128 pages

Only entire blocks may be written or rased on a solid state memory chip.

Front 55

Page States

Back 55

Erased (empty)

Valid

Invalid

- In order to write any data to a page, its owning block location on the flash memory chip must be electrically erased.

Front 56

Garage collection

Back 56

The process of providing new erased Blocks

- When an invalid page needs to be erased before it can once again be written with new data.

Front 57

Block States

Back 57

Erased (Empty)

New

Used

Front 58

Write Request- Invalidation of a page

Back 58

If the drive receives a write request to a valid block page, the page must be changed. The current page containing the destination of the write is marked invalid. The block's state changes to "used because it contains invalid pages.

Front 59

Delete Request - Invalidation of a page

Back 59

A delete invalidates a page without resulting in a subsequent write

Front 60

Access Type of SSDs

Back 60

- SSDs perform random reads the best

- SSDs use all internal I/O in parallel for multi-threaded large blocks I/Os

Front 61

Drive State of an SSD

Back 61

New SSD or SSD with substantial unused capacity offers the best performance

Front 62

SSD Workload duration

Back 62

SSDs are best for workloads with short burst of activity

Front 63

Why RAID?

Back 63

RAID--> Redundant Array of Independent Disks

A technique that combines multiple disk drives into a logical unit (RAID set) and provides protection, performance, or both.

Front 64

Benefits of RAID

Back 64

- Provides data protection against drive failures

- Improves storage system performance by serving I/Os from multiple drives simultaneously

Front 65

Key Function of RAID

Back 65

- Management and control of drive aggregations

- Translation of I/O requests between logical and physical drives, and data regeneration in the event of drive failures.

Front 66

Software RAID

Back 66

Uses compute system based software to provide RAID functions and is implemented at the operating system level.

PROS- cost and simplicity benefits when compared to hardware RAID

Front 67

Limitations of Software RAID

Back 67

- Performance (Additional CPU cycles)

- Supported Features (Does not support all RAID levels)

- Upgrades to software ROAD or to the OS must be validate and this leads to inflexibility in the data-processing environment.

Front 68

Key Functions of RAID Controllers

Back 68

- Management and control of disk aggregations

- Translation of I/O between logical disk and physical disks

-Data regeneration in the event of disk failures

Front 69

Controller Card RAID

Back 69

Compute-system based hardware RAID Implementation which a specialized RAID controller is installed in the compute system, and disk drives are connected to it.

Front 70

External RAID Controller

Back 70

A storage-system based hardware RAID

- Acts as an interface between the compute system and the disks. It presented storage volume to compute system, and the compute system manages the volumes as physical drives.

Front 71

Raid Array

Back 71

An enclosure that contains a number of disk drives and supporting hardware to implement RAID

Front 72

RAID Group

Back 72

A subset of disks within a RAID array can be grouped to form logical associations called logical arrays known as a RAID Set/ RAID Group.

Front 73

Striping

Back 73

A technique of spreading data across multiple drives (more than one) in order to user the drives in parallel.

Front 74

Strip Size

Back 74

Describes the number of blocks in a strip, and is the maximum amount of data starts at the beginning of the strip.

Also called Stripe Depth

Front 75

Strip Width

Back 75

Refers to the number of data strips in a stripe.

Front 76

Mirroring

Back 76

A technique whereby the same data is stored on two different disk, yielding two copies of the data.

Front 77

Parity

Back 77

A method to protect striped data from disk drive failure without the cost of mirroring.

Front 78

PROs of Mirroring

Back 78

- Preferred for mission-critical applications that cannot afford the risk of any data loss.

- Improves read performance because read quests can be serviced by both disks.

Front 79

PROs of Parity

Back 79

-Ensures protection of data without maintaining a full set of duplicated data

-Parity calculation is a bitwise XOR operation.

- Considerably reduces the cost associated with data protection

Front 80

RAID Level - Unprotected

Back 80

No Raid, 1 Disk, "Because I'm Stupid"

Front 81

RAID Level 0

Back 81

Striping

- Good option for applications that need high I/O throughput

Minimum number of disks: 2*

EMC Prefers (3-16)

Front 82

RAID Level 1

Back 82

Mirroring

Minimum number of disks: 2

- Suitable for application that require high availability and cost is not a constraint

Write Penalty: 2

Available Storage Capacity 50%

Front 83

RAID Level 3

Back 83

Striped set with parallel access and dedicated parity disk

Minimum number of disks: 3* (2+1)

EMC Prefers: (5 or 9) (4+1, 8+1)

Available Storage Capacity: [(n-1)/n]*100

Front 84

RAID Level 5

Back 84

Striped set with independent disk access and a distributed parity

Minimum number of disks:

VMAX: 3+1 or 7+1

VNX: 3-16

Available Storage Capacity: [(n-1)/n]*100

Front 85

RAID Level 6

Back 85

Striped with independent disk access and dual distributed parity.

Minimum number of disks: 4 (2+2)

EMC Symme. (6+2, 14+2)

Available Storage Capacity: [(n-2)/n]*100

Front 86

RAID Level 1/0

Back 86

Nested RAID- known as RAID 10(Ten)

Mirroring THEN Striping

Minimum number of disks: 4

Write Penalty: 2

Available Storage Capacity 50%

Front 87

VMAX

Back 87

Local: Timefinder

Remote: SRDF

Front 88

VNX

Back 88

Local: Snap view

Remote: Mirrorview

Front 89

Hot Sparing

Back 89

A process that temporarily replaces a failed disk drive with a spare drive in a RAID array by taking the identity of the failed disk drive.

- IF parity is used, the data is rebuilt onto the hotspare from the parity and the data on surviving disk drives in the RAID set.

-IF mirroring is used, the data from the surviving mirror use used to copy the data onto the hot spare.

Front 90

Block-level Access

Back 90

The file system is created on a compute system, and data is accessed on a network at the block level. Raw disks or logical volumes are assigned to the compute system for creating the file system.

Front 91

File-level Access

Back 91

The file system is created on a separate file service or at the storage side, and the file-level request is sent over a network. Because data is access at the file level, this method has higher overhead, when compared to block.

Front 92

Object-level Access

Back 92

Data is accessed over a network in terms of self-contained objects with a unique object identifier.

The files system's user component resides on a the computer system and the storage component resides on the storage system.

Front 93

Scale-Up

Back 93

Architecture that provides the capability to scale the capacity and performance of a single storage system.

- This involves upgrading or adding controller and storage.

Front 94

Scale-Out

Back 94

Architecture that provides capability to maximize its capacity by simply adding nodes to the cluster.

- Nodes are quick to add when more performance and capacity are needed without downtime.

- Distribution of workload across all the nodes.