Storage Intro Flashcards
(105 cards)
1
Q
Within the world of IT, along with Computing and Networking what is the other major IT infrastructures?
A
Storage
2
Q
What resides in the Storage layer of the infrastructure?
A
All Data
3
Q
What are the 2 high level types of storage?
A
Persistent(non-volatile) and Nonpersistent (volatile)
4
Q
What is the standard choice for long term storage of data?
A
Persistent storage
5
Q
Why is long term storage called Persistent/nonvolatile?
A
Does not loose it’s content when the power is turned off
6
Q
What is the most common example of nonpersistent/volatile storage?
A
RAM (Random access memory)
7
Q
What are the two important characteristics of both mechanical and solid state storage device?
A
Performance and availability
8
Q
What are the three dominate storage devices?
A
Disk, Solid-state, and Tape
9
Q
Disk, Solid-state and Tape are three forms of what?
A
Storage and media
10
Q
Electromechanical hard disk drive refers to which type of device?
A
Disk storage
11
Q
Which type of media refers to the newer technologies in the data centers?
A
Solid-state media
12
Q
Disk, solid-state and tape storage are all forms of what kind of storage?
A
Persistent/nonvolatile storage
13
Q
What is Crash Consistent?
A
maintaining the correct order of write in order to allow an application to restart properly from a crash.
14
Q
What are the 4 major mechanical components of the disk drive?
A
Platters, Read/write heads, Actuator assembly and spindle motor
15
Q
On a mechanical disk drive, where is the data stored?
A
Platter
16
Q
What is the mechanical part of the disk drive that read and write data to the platters?
A
Read/Write heads (R/W head)
They are controlled by the firmware on the disk drive controller.
17
Q
The flying height of the read/write heads are measured in what?
A
Nanometers
18
Q
What is known as a head crash?
A
when the read/write heads touch the platters
19
Q
What is CHS?
A
Cylinder head sector ( it is an addressing scheme)
20
Q
What are the three things used to address a sector in a disk drive?
A
Cylinder number (gives us the track), head number (tells which recording surface the track is on) and sector number (tells which sector on the track we have just identified).
21
Q
Platters are microscopically divided into what?
A
Tracks and Sectors
22
Q
What is the smallest addressable unit of a disk drive?
A
The sector (typically 512 or 520 bytes in size)
23
Q
What is EDP?
A
End to end Data Protection
24
Q
What is End to end Data Protection?
A
Allows drives to detect errors either before committing data to the drive or before returning corrupted data to the host/application.
25
EDP adds 8 bytes of data which includes a CRC (cyclic redundancy check) that allows the checking of intergrity is called what?
Guard field
26
What is ZDR?
Zoned data recording ( used to make better use of available space on a disk)
27
What is the purpose of a servo signal?
They assist in keeping read/write heads on track on adjacent sectors
28
What is FBA?
Fixed block architecture (it is where the sector size is preset and can't be changed).
29
What is a cylinder?
a collection of tracks stacked directly above one another on separate platters
30
What is LBA?
Logical block addressing ( scheme used for specifying the location of blocks of data on a computer storage device.
31
What owns the LBA map?
Drive controller
32
What is the function of the Actuator assembly?
to physically move the R/W heads (under the direction of the disk firmware and controller).
33
What is the disk drive controller?
it is like a mini computer with it's own processor and memory, and it runs the firmware that is vital to the operation of the drive.
34
What are some of the functions of the controller?
Monitors the health of the drive, reports potential issues and maintains the bad block map.
35
What are the four most common protocols and interfaces in the drive world?
SATA (Serial Advanced Technology Attachment), SAS (Serial Attached SCSI), NL-SAS (Nearline SAS) and FC (Fibre Channel)
36
What is Latency?
Seek time and rotational delay
37
In the Enterprise tech, SATA drives are known for what?
Cheap, low performance and high capacity.
38
A serial point to point protocol that uses the SCSI command set is know as what?
SAS (Serial Attached SCSI)
39
The best choices for high performance and mission critical workloads are?
SAS and FC, because of their SCSI command set.
40
What is Metadata?
A set of data that describes and gives information about the other data.
41
Which drive interface is a blend of SAS and SATA?
NL-SAS
42
What is queuing?
Allows the drive to reorder I/O operations so that the read/write commands are executed in an order optimized for the layout of the disk.
43
What is RPM?
Revolutions per minute
44
What 2 major factors influence disk drive performance?
Seek Time and Rotational Latency
45
What is positional latency?
Anything that requires the heads to move or wait for the platter to spin into position.
46
What is Seek Time?
The time taken to move the read/write heads to the correct position on the disk.
47
Seek Time is expressed in what?
Millisecond (ms) (one-thousandth of a second)
48
What is rotational Latency?
The time is takes for the correct sector to arrive under the R/W head after the head is positioned on the right track,
49
What does IOPS stand for?
input/output operations per second
50
What is IOPS used for?
To measure and express the performance of mechanical, solid-state and storage arrays
51
What is an I/O operation?
a read or write operation that a disk or disk array performs in response to a request from a host (usually a server)
52
What are the most important I/O operation?
Read, Write, random, Sequential, Cache hit, and Cache miss
53
What is Cache-hit IOPS?
Referred to Cache IOPS, are IOPS satisfied from cache rather than disk.
54
What is the formula for calculating how many IOPS a disk drive should be able to service?
1/(X+Y)1,000: X=average seek time and Y= Average rotational latency
55
What is MBps?
Megabytes per second
56
What is Maximum Transfer Rate?
in reference to disk drive performance, it is used to express the absolute highest possible transfer rate of a disk drive under optimal conditions
57
What is STR (Sustained transfer rate?
the rate at which a disk drive can read or write sequential data spread over multiple tracks
58
What is MTBF?
Mean time between failure
59
What is AFR?
Annual failure rate ( this attempts to estimate the likelihood that a disk drive will fail during a year of full use.
60
What is SSD?
Solid State Drive and it has no mechanical parts. They are silicon/semiconductor based,
61
What is Flash Memory?
It is a form of solid-state storage. No mechanical parts, semiconductor based and provides persistent storage.
62
What are the 4 major types of NAND flash?
Single-level cell, Multi-level cell, Enterprise MLC, and triple-level cell
63
What happens in a SLC (Single-level Cell) world?
a single flash cell can store 1 bit of data; the electrical charge applied to the cell could be either on or off (so you can have either binary 0 or binary 1).
64
Which flash cell option provides the highest performance and the longest life, but the most expensive one?
SLC
65
What is Multi-level cell (MLC)?
stores 2 bits of data per flash cell (00, 01, 10, or 11), allowing it to store twice as much data per flash cell as SLC does.
66
What is Enterprise-grade MLC flash (eMLC)?
flash, known more commonly as eMLC, is a form of MLC flash memory with stronger error correction, and therefore lower error rates, than standard MLC flash.
67
When does Write Amplification occur?
occurs when the number of writes to media is higher than the number of writes issued by the host . It slows flash performace and unnecessarily wears out the flash cells.
68
What is Write Endurance?
refers to the number of program/ erase (P/ E) cycles a flash cell is rated at.
69
what is Wear-Leveling?
is the process of distributing the workload across all flash cells so that some don’t remain
youthful and spritely while others become old and haggard.
70
What is In-Flight wear leveling?
silently redirects incoming writes to unused areas of flash that have been subject to lower P/ E cycles than the original target cells. The aim is to make sure that some cells aren’t unduly subject to higher numbers of P/ E cycles than others.
71
What is Background wear leveling?
is a controller-based housekeeping task. This process takes care of the flash cells that contain user data that doesn’t change very often (these cells are usually off-limits to in-flight wear leveling).
72
What is Write coalescing?
is the process of holding writes in cache until a full page or block can be written.
73
What is Write-combining?
happens when a particular LBA address is written to over and over again in a short period of time .
74
What is Caching?
is as simple as putting a small amount of memory in front of disk so that frequently accessed data can be accesses from cache memory rather than having to go to disk.
75
What is S.M.A.R.T?
Self-monitoring, Analysis, and Reporting Technology
76
What is the purpose of S.M.A.R.T?
It is an industry-wide monitoring standard for hard disk drives that attempts to predict failures.
77
When does filesystem fragmentation occur?
when files do not occupy contiguous addresses within the filesystem.
78
When does disk fragmentation occur?
When a file, which might be totally unfragmented in the filesystem, is fragmented on the disk.
79
What is a Storage array?
A computer system designed for and dedicated to providing storage to externally attached computers usually via a storage network.
80
What are the 3 major Storage arrays?
SAN, NAS, and Unified (SAN and NAS)
81
What is SAN storage array?
Sometimes referred to as Block storage array, Provide connectivity via block-based protocols. (FC-Fibre Channel, FCoE- Fibre Channel over Ethernet, ISCSI- Internet Small Computer System interface, or SCSI-Serial attached SCSI-SAS
82
What does block storage array send?
low-level disk-drive access commands called SCSI command descriptor blocks (CDBs) such as READ block, WRITE block, and READ CAPACITY over the SAN.
83
What is NAS storage array?
sometimes called filers, provide connectivity over file-based protocols such as Network File System (NFS) and SMB /CIFS.
.
84
Where do file-based protocols work?
at a higher level than low-level block commands.
85
What do file-based protocols do?
They manipulate files and directories with commands that do things such as create files, rename files, lock a byte range within a file, close a file, and so on.
86
What is CIFS?
Common Internet File System
87
What is Unified storage arrays?
sometimes referred to as multiprotocol arrays, provide shared storage over both block and file protocols.
88
what is the purpose of all storage arrays?
is to pool together storage resources and make those resources available to hosts connected over the storage network.
89
How are storage resources presented to hosts?
as SCSI logical unit numbers (LUNs).
90
What is a LUN?
looks and behaves exactly like a locally installed disk drive— basically, a chunk of raw capacity.
91
What is Redundancy in IT design?
is the principle of having more than one of every component so that when (not if) you experience component failures, your systems are able to stay up and continue providing service.
92
What are Dual-Controller Architectures?
Storage arrays with 2 controllers
93
What is Grid storage architectures?
sometimes referred to as clustered or scale-out architectures, are the answer to the limitations of dual-controller architectures. They consist of more than two controllers and are far more modern and better suited to today’s requirements than legacy dual-controller architectures.
94
What is the main purpose of cache?
to hide the mechanical latencies of spinning disk
95
What is all-flash storage array?
a solid state storage disk system that contains multiple flash memory drives instead of spinning hard disk drives
96
What is deduplication?
a specialized data compression technique for eliminating duplicate copies of repeating data
97
What do Storage arrays allow you to do?
to pool storage resources, thereby making more-efficient use of both capacity and performance
98
What is Multipath I/O (MPIO) Software?
controls how data is routed or load balanced across these multiple links, as well as taking care of seamlessly dealing with failed or flapping links.
99
What are some common MPIO load balancing algorithms?
Failover only, Round Robin and Least Queue Depth
100
What is Failover Only?
Where one path to a LUN is active and the other passive, no load balancing is performed across the multiple paths.
101
What is Round robin?
I/ O is alternated over all paths.
102
What is Least Queue Depth?
The path with the least number of outstanding I/ Os will be used for the next I/ O.
103
What is LUN Masking?
is the process of controlling which servers can see which LUNs.
104
What is Referential locality?
refers to how widely spread over your address space your data is.
105
What is Prefetching?
a common technique that arrays use when they detect sequential workloads
