0% found this document useful (0 votes)
56 views16 pages

Information Storage System-Chapter2

The document summarizes different types of storage interfaces and protocols including SATA, SAS, FC, and NL-SAS. It discusses how disk drive size, capacity, performance, and speed are measured and expressed. Key points covered include seek time, rotational latency, RPM, command queuing techniques like NCQ and CQ, and factors that reduce the usable capacity of disk drives compared to their advertised size.

Uploaded by

Salha Bujazia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
56 views16 pages

Information Storage System-Chapter2

The document summarizes different types of storage interfaces and protocols including SATA, SAS, FC, and NL-SAS. It discusses how disk drive size, capacity, performance, and speed are measured and expressed. Key points covered include seek time, rotational latency, RPM, command queuing techniques like NCQ and CQ, and factors that reduce the usable capacity of disk drives compared to their advertised size.

Uploaded by

Salha Bujazia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Information Storage

Systems
LIMU
SPRING 2017-2018
Objectives
List the different types of interfaces and protocols in storage networks.
Explain how disk drive size is expressed.
Explain how drive disk performance is measured and expressed.
Serial Advanced Technology Attachment
The underlying command set for SATA is ATA.
In enterprise tech, SATA drives are synonymous with cheap, low performance, and high capacity.
The major reason for this is that the ATA command set is not as rich as the SCSI command set,
and as a result not as well suited to high-performance workloads.
However, the enterprise tech world is not all about high performance. There is absolutely a
place for low-cost, low-performance, high-capacity disk drives. Some examples are backup and
archiving appliances, as well as occupying the lower tiers of auto tiering solutions in storage
arrays.
SATA is being replaced in storage arrays by NL-SAS.
Serial Advanced Technology Attachment
Serial Attached SCSI
Small Computer System Interface
SCSI has its roots family in the high-end enterprise tech world, leading to it have a richer command set,
better queuing system, and often better physical component quality than SATA drives.
All this tends to make SCSI-based drives, such as SAS and FC, the best choice for high-performance,
mission-critical workloads.
Another key advantage of SAS is that SATA II, and newer, drives can connect to a SAS network or
backplane and live side by side with SAS drives. This makes SAS a flexible option when building storage
arrays.
SAS drives are also dual ported (Which is useful for redundancy) each port on the SAS drive can be
attached to different controllers in the external storage array.
These ports work in an active/passive mode, meaning that only one port is active and issuing
commands to the drive at any one point in time.
Serial Attached SCSI
Finally, SAS drives can be formatted with
arbitrary sector sizes, allowing them to
easily implement end to end protection
(EDP) sometimes called Data Integrity Field
(DIF). With EDP, 520-bye sectors are used
instead of 512-bye sectors. The additional 8
bytes per sector are used to store metadata
that can ensure the integrity of the data,
making sure that it hasn’t become
corrupted.
Fibre Channel
FC drives implement the SCSI command set and queuing just like SAS, suited to high-
performance requirements and come at a relatively high cost per gigabyte, but are very much
being superseded by SAS drives.
NL-SAS
Nearline SAS (NL-SAS) drives are a hybrid of SAS and SATA drives. They have a SAS interface and
speak the SAS protocol, but also have the platters and RPM of a SATA drive.
they slot easily onto a SAS backplane or connector and provide the benefits of the SCSI
command set and advanced queuing, while at the same time offering the large capacities
common to SATA.
NL-SAS has all but seen off SATA in the enterprise storage world, with all the major array vendors
supporting NL-SAS in their arrays.
Queuing
All disk drives implement a technique known as queuing. Queuing has a positive impact on
performance.
Queuing allows the drive to reorder I/O operations so that the read and write commands are
executed in an order optimized for the layout of the disk. This usually results in I/O operations
being reordered so that the read and write commands can be as sequential as possible so as to
inflict the least head movement and rotational latency as possible.
The ATA command set implements Native Command Queuing (NCQ). This improves the
performance of the SATA drive, especially in concurrent IOPS, but is not as powerful as its cousin
Command Tag Queuing in the SCSI world.
Drive Size
When speaking of disk drives, there are two types of size you need to be aware of:
Capacity
Physical form factor
Normally, when referring to the size of a drive, you are referring to its capacity, or how much
data it can store.
However, the size of a drive can also refer to the diameter of the drive’s platter, which is
normally referred to as the drive’s form factor. In modern disk drives, there are two popular HDD
form factors:
3.5-inch
2.5-inch
Drive Size
All 3.5-inch drives come in the same-size casing, meaning that all 3.5-inch drives can fit in the same
slots in a server or disk array. The same goes for 2.5-inch drives. All 2.5-inch drives come in the same-
size casing, meaning that any 2.5-inch drive can be swapped out and replaced by another. So even
though technically speaking, 2.5-inch and 3.5-inch refer to the diameter of the platter, more often
than not when we refer to the physical dimensions of one of these drives, we are referring to the size
of the casing that it comes in. Table below shows the measurements of each of these two form
factors (Same goes to SSD)
Usable Capacity
While speaking about capacity, it is worth pointing out that you don’t typically get what it says on the
tin when you buy a disk drive. For example, when buying a 900 GB drive, you almost certainly won’t
get 900 GB of usable space from that drive. There are several reasons for this.
One reason is that storage capacities can be expressed as either base 10 (decimal) or base 2 (binary).
As an example, 1 GB is slightly different in base 10 than in base 2:
Base 10 = 1,000,000 bytes
Base 2 = 1,048,576 bytes
Other factors play into this as well. For example, when formatting a disk in an OS, the OS and
filesystem often consume some space as overhead. Similarly, when installing a disk in a storage array,
the array will normally shave off some capacity as overhead.
There is also RAID and other things to consider. The bottom line is, don’t be surprised if you get slightly
less than you bargained for.
Drive Speed
Drive speed is all about how fast the platters spin—and they spin really fast (5.4 k , 7.2k, 10k, 15k).
If RPM doesn’t do it for you, look at it like this: a 3.5-inch platter spinning at 15,000 RPM is about 240
kilometers per hour.
RPM converts directly into drive performance. More RPMs = more performance (and higher price, of course).
There is also a strong correlation between RPM and capacity. Usually, the higher the RPM of a drive, the lower
the capacity of that drive. Conversely, the lower the RPM of a drive, the higher the capacity.
Another point worth noting is that 10K and 15K drives usually implement the SCSI command set and have a
SAS or FC interface. So, faster drives implement the more performance-tuned protocol and interface. That
makes sense. Also, the 5.4K and 7.2K drives usually implement the ATA command set and have a SATA
interface.
2.5-inch, 900 GB, 10K SAS
3.5-inch, 4 TB, 7,200 RPM SATA
Disk Drive Performance
Several factors influence disk drive performance, and the following sections cover the more
important ones. We’ve already touched on some of these, but we will go in to more detail now.
Two major factors influence disk drive performance:
Seek time
Rotational latency
Both of these fall into the category of positional latency (anything that requires the heads to
move or wait for the platter to spin into position).
Seek Time
Seek time is the time taken to move the read/write heads to the correct position on disk (positioning
them over the correct track and sector). Faster is better, and as seek time is expressed in milliseconds
(ms), the lower the number, the better.
Seek time can therefore have a significant impact on random workloads, and excessive seeking is often
referred to as head thrashing or sometimes just thrashing. And thrashing absolutely kills disk drive
performance.
On very random, small-block workloads, it is not uncommon for the disk to spend over 90 percent of its
time seeking. As you can imagine, this seriously slows things down! For example, a disk may achieve
along the lines of 100 MBps with a nice sequential workload, but give it a nasty random workload, and
you could easily be down to less than a single MBps!
Some real-world average seek times taken from the spec sheets of different RPM drives are listed here:
(In this list, the rst number on each line is for reads, and the second number is for writes.)
15K drive: 3.4/3.9 ms
7.2K drive: 8.5/9.5 ms
Rotational Latency
Rotational latency is the other major form of positional latency. It is the time it takes for the
correct sector to arrive under the R/W head after the head is positioned on the right track.
Rotational latency is directly linked to the RPM of a drive. Faster RPM drives have better
rotational latency (that is, they have less rotational latency). While seek time is significant in only
random workloads, rotational latency is influential in both random and sequential workloads.
Disk drive vendors typically list rotational latency as average latency and express it in
milliseconds. Average latency is accepted in the industry as the time it takes for the platter to
make a half rotation (that is, spin halfway around).
15K drive = 2.0 ms
7.2K drive = 4.16 ms

You might also like