Talk:Hard disk drive performance characteristics

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing / Hardware (Rated Start-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
Taskforce icon
This article is supported by Computer hardware task force (marked as Mid-importance).
 
edit·history·watch·refresh Stock post message.svg To-do list for Hard disk drive performance characteristics:

To-do list is empty: remove {{to do}} tag or click on edit to add an item.


Leave overall power consumption section in HDD article[edit]

Proposed we keep the overall power consumption section in HDD article. I pulled out the key elements and included them in the article, but so much of this is great data on power which I believe should stay with the HDD article. At this time I have not removed it from the HDD article.


Heat dissipation is tied directly to power consumption, and as drives age, disk [[failure rate]]s increase at higher drive temperatures.<ref name="xbit-2007-12-06">{{cite news |newspaper=Xbit Laboratories |title=Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology |date=6 December 2007 |first=Oleg |last=Artamonov |url=http://www.xbitlabs.com/articles/storage/display/hdd-power-cons.html |accessdate=2011-07-03}}</ref> Similar issues exist for large companies with thousands of desktop PCs. Smaller form factor drives often use less power than larger drives. One interesting development in this area is actively controlling the seek speed so that the head arrives at its destination only just in time to read the sector, rather than arriving as quickly as possible and then having to wait for the sector to come around (i.e. the rotational latency).<ref>e.g. Western Digital's [http://www.wdc.com/en/flash/index.asp?family=intelliseek Intelliseek]</ref> Many of the hard drive companies are now producing Green Drives that require much less power and cooling. Many of these Green Drives spin slower (<5,400 rpm compared to 7,200, 10,000 or 15,000 rpm) thereby generating less heat. Power consumption can also be reduced by parking the drive heads when the disk is not in use reducing friction, adjusting spin speeds,<ref>[http://www.xbitlabs.com/news/storage/display/20071022123416.html Hitachi Unveils Energy-Efficient Hard Drive with Variable Spindle Speed.]</ref> and disabling internal components when not in use.<ref>{{cite book |title=Green tech: how to plan and implement sustainable IT solutions |first=Lawrence |last=Webber |first2=Michael |last2=Wallace |isbn=081441446X |year=2009 |url=http://books.google.com/books?id=BKTALNq5ceAC&lpg=PA62&dq=green%20disk%20drive&pg=PA62#v=onepage&q=green%20disk%20drive&f=false |page=62 |accessdate=2011-07-03}}</ref> Also in systems where there might be multiple hard disk drives, there are various ways of controlling when the hard drives spin up since the highest current is drawn at that time. *On SCSI hard disk drives, the SCSI controller can directly control spin up and spin down of the drives. *On [[Parallel ATA]] (aka PATA) and [[Serial ATA]] (SATA) hard disk drives, some support [[power-up in standby]] or PUIS. The hard disk drive will not spin up until the controller or system BIOS issues a specific command to do so. This limits the power draw or consumption upon power on. *Some SATA II hard disk drives support staggered spin-up, allowing the computer to spin up the drives in sequence to reduce load on the power supply when booting.<ref>{{cite news |title=Hitachi Deskstar 7K500 500GB HDD: As fast as it's big? |author=Trusted Reviews |date=31 August 2005 |http://www.theregister.co.uk/2005/08/31/review_hitachi_7k500/ |accessdate=2011-07-03}}</ref> Most hard disk drives today support some form of power management which uses a number of specific power modes that save energy by reducing performance. When implemented an HDD will change between a full power mode to one or more power saving modes as a function of drive usage. Recovery from the deepest mode, typically called Sleep, may take as long as several seconds.<ref>[http://www.almaden.ibm.com/almaden/mobile_hard_drives.html#2 Adaptive Power Management for Mobile Hard Drives]</ref> ====Shock resistance==== Shock resistance is especially important for mobile devices. Some laptops now include [[active hard drive protection]] that parks the disk heads if the machine is dropped, hopefully before impact, to offer the greatest possible chance of survival in such an event. Maximum shock tolerance to date is 350 [[Gravitational acceleration|g]] for operating and 1000 g for non-operating.<ref>[http://www.seagate.com/ww/v/index.jsp?vgnextoid=5fb658a3fd20a110VgnVCM100000f5ee0a0aRCRD Momentus 5400.5 SATA 3Gb/s 320-GB Hard Drive]</ref>

§ Music Sorter § (talk) 07:33, 5 July 2011 (UTC)

Interleave removed from combined articles[edit]

[[Image:IBM PC XT 10 meg MFM low level format.jpg|thumb|200px|A low-level formatting tool running tests to find the highest performance interleave choice for a 10-megabyte IBM PC XT hard drive.]] Sector interleave is a mostly obsolete device characteristic related to access time, dating back to when computers were too slow to be able to read large continuous streams of data. Interleaving introduced gaps between data sectors to allow time for slow equipment to get ready to read the next block of data. Without interleaving, the next logical sector would arrive at the read/write head before the equipment was ready, requiring the system to wait for another complete disk revolution before reading could be performed.

However, because interleaving introduces intentional physical delays into the drive mechanism, setting the interleave to a ratio higher than required causes unnecessary delays for equipment that has the performance needed to read sectors more quickly. The interleaving ratio was therefore usually chosen by the end-user to suit their particular computer system's performance capabilities when the drive was first installed in their system.

Modern technology is capable of reading data as fast as it can be obtained from the spinning platters, so hard drives usually have a fixed sector interleave ratio of 1:1, which is effectively no interleaving being used.

Access time merger[edit]

Access time is common to all data storage and the disk specific issues are all ready well covered in this article and/or its parent. Arguably it applies to data communications also (at least latency and data rate do), although I don't recall the term being used too much in that art. So there is nothing to merge here. Rather the Access Time article needs a lot of work to broaden it or we might merge it into Computer data storage. Tom94022 (talk) 15:28, 9 July 2011 (UTC)

I certainly agree that we have the current disk drive elements of access time covered, but I did not want to lose the data comm element of access time. I had not investigated whether the data communication "access time" was legitimate or not, but either way I figured we could either disambiguate it to separate the two and send the drive related readers here, or add "other uses" to the access time article. I am open to either. § Music Sorter § (talk) 16:46, 9 July 2011 (UTC)

Note: the following was placed on the talk:Access time page and I copied it here for simplicity of seeing all the feedback. § Music Sorter § (talk) 15:16, 12 July 2011 (UTC)

Agree Merge the disk-related section to Disk drive performance characteristics, simply because there it would contribute to an interesting article whereas here it can say nothing beyond a simple, unhelpful definition. --Kubanczyk (talk) 09:10, 12 July 2011 (UTC)

  • Dabify. In addition to disk access, Unix file stats is certainly valid, and possibly telecom, too. Support a selective merge of the disk-related content into Disk drive performance characteristics in agreement with everyone who posted above. --Pnm (talk) 03:31, 1 August 2011 (UTC).
  • Merge Just makes sense to merge, access time is a part of performance,0pen$0urce (talk) 17:29, 7 October 2011 (UTC)

Seek time merger[edit]

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
After 15 days, the result of this discussion was to merge the content as proposed with no opposition to the one nomination to move it. § Music Sorter § (talk) 07:56, 26 July 2011 (UTC)

All of the content in the current stub Seek time article has been included in this article. I propose we change this page to a redirect to Disk drive performance characteristics#Seek time. § Music Sorter § (talk) 05:33, 10 July 2011 (UTC)


The above discussion is preserved as an archive. Please do not modify it. Subsequent comments should be made in a new section.


Rotational delay merger[edit]

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
The result of this discussion was to merge Rotational delay into Disk drive performance characteristics#Rotational latency. § Music Sorter § (talk) 06:42, 16 July 2011 (UTC)

All of the content in the current stub Rotational delay article has been included in this article. I propose we change this page to a redirect to Disk drive performance characteristics#Rotational delay.§ Music Sorter § (talk) 07:43, 10 July 2011 (UTC)

Support. -- intgr [talk] 13:16, 10 July 2011 (UTC)
Support as proposer. § Music Sorter § (talk) 03:19, 11 July 2011 (UTC)
Support. -- jbc (talk) 11:13, 11 July 2011 (UTC)
Sure. But only if you accept that Access and Seek times are also kinds of latency. Currently, you associate the disk latency with rotational performance only. --Javalenok (talk) 13:59, 11 July 2011 (UTC)
Javalenok, the current Rotational Delay article includes a sentence "Rotational delay is one of the three delays associated with reading or writing data on a disk, and somewhat similar for CD or DVD drives. The others are seek time and transfer time." This sentence says that there are three things that can delay a drive: rotational delay (latency), seek time, and transfer time. It does not say that they are kinds of latency. I have not seen seek time and transfer time referenced as latency with respect to disk drives in the past. Do you have a source for your statement? § Music Sorter § (talk) 15:31, 11 July 2011 (UTC)
Common sense. I know it is considered inappropriate in the Wikipedia if we forget that statements like "water is wet" are allowed without a reference. --Javalenok (talk) 15:12, 13 July 2011 (UTC)
Music Sorter, I assume most readers would consider 'delay' and 'latency' to be synonymous in this context. Wiktionary defines latency as "A delay, a period between the initiation of something and the occurrence". Ditto the Wikipedia latency (engineering) article. Trying to distinguish between these words will only confuse readers. I'd prefer if you always qualify it as "rotational latency" instead of "latency". -- intgr [talk] 15:59, 13 July 2011 (UTC)
If I am wrong, I am in full support of your descriptions. I don't want to be argumentative for its own sake, so I will only see if I can present the other side of the sources. After that I am fine with consensus. I agree that the general definition for those terms are often used interchangeably, but that is part of the reason so many people misunderstand the terminology. Isn't that why we are here editing these pages? Don't we want to ensure the more accurate information supported by the most qualified sources? Following are some examples of sources with disk drive backgrounds or references that separate the term Latency to only reference the period resulting from the rotation of the platters and not the Seek time or transfer time.
  • SearchStorage.co.uk "...Seeks and latency. How quickly the disk can find and read a sector is determined in part by access time. Reading a particular sector consists of two steps. First, the head must be moved to the correct track. Then, once the head is over that track, you must wait for the sector to spin under the head and read the sector. Seek time is the time required for the head to position itself over a track. The latency period is how long it takes the desired sector to move under the head." Seek and Latency are two separate terms related to access time.
  • IBM.com Developerworks "Characteristics that affect HDD performance...Actuator Positioning (Seek Time). This is the time it takes the actuating mechanism to move the heads from their present position to a new position. Seek time typically averages a few milliseconds and is dependent upon the drive type. For example, a 15,000 RPM drive has an average seek time of approximately 3.5 milliseconds for reads and 4 milliseconds for writes. (The full disk seek time is 7.4 milliseconds for reads and 7.9 milliseconds for writes...Rotational latency. This is the time it takes to correctly position the platter underneath the head so that the desired data can be accessed. (The average rotational latency is one-half of a revolution of the disk platter.) In the case of a 15,000 RPM drive, this is approximately 2 milliseconds." IBM defines the positioning of the actuator as seek time and the rotational latency as waiting for the platter under the head to get into position.
  • New York Data Recovery (Disk Drives) "Latency - The period of time that the read/write heads wait for the disk to rotate to the correct position to access the desired data. For a disk rotating at 5200 RPM, the average latency is 5.8 milliseconds; or, the average time delay between the head arriving on track and the data rotating to the head. (Calculated as one-half the revolution period.) Seek Time - A measure (in milliseconds) of how fast the hard drive can move its read/write heads to a desired location. Access Time - The amount of time, including seek time, latency and controller time, needed for a storage device to retrieve information." You can see they define Access time as three separate factors distinct from each other.
  • Seagate Glossary Latency The time for the disc to rotate the accessed sector under the head for read or write. On the average, latency is the time for half of a disc revolution.
  • Seagate Glossary Seek time The amount of time it takes the Read/Write heads to travel from their current cylinder location to a new cylinder and includes head settling time.
intgr, in the world of disk drives, all latencies are a form of delay, but not all delays are a form of latency.
Javalenok, yes, all water is wet, but everything wet is not water.
Proposal: Presuming the data here as presented provided a compelling justification for not merging the term latency and delay, I fully support intgr's proposal to change the Disk drive performance characteristics#Latency section heading to Rotational latency instead. Further, I can add some detail in the newly renamed Rotational latency section to discuss how in some engineering and computing circles, the term latency is often used synonymously with delay, but in disk drive terminology all latencies can cause a delay in time, but not all delays in time come from latencies. § Music Sorter § (talk) 07:11, 14 July 2011 (UTC)
Curious people always try to understand the difference between the terms. Why one thing is used here and another there? The terminology problem emerges when somebody starts using the same word for different things or refer to the same thing with different words. Your baggage becomes "a luggage" when you move from blue train to a green one. It is opposite if you travel by buses. Others decide to promote such terminology in Wikipedia and, thereby, reenforce it. The zeal becomes inventive when asked about the difference between baggage and luggage: "One is a subclass of another"! It is a nonsense of course. I suspect that the HDD manufacturers merely have to deal with both latencies and optimized their communication this way. May be, it just sounds better in English. "Red bus transported the baggage." Sounds terrible. Ugh. --Javalenok (talk) 09:36, 18 July 2011 (UTC)
Well you've certainly done your homework. :) I guess the terminology is quite clear when you read resources exclusively devoted to hard drives. But as you mention, people from other fields frequently confuse the terms "latency" and "delay". Hell, even the Wikipedia article is called rotational delay not rotational latency. :)
I agree with your proposal, "rotational latency" is correct and unambiguous. -- intgr [talk] 11:49, 14 July 2011 (UTC)
Do you say that terminology is "Correct and unambigous" because it contradicts to the common sense and, thus causes countless confusion as you prove in your examples? With your phrase "water is a subclass of wet substances" you disregard my point that common sense is allowed to use in Wikipedia. Instead, you put forward your idea that the latency is a special case of delay. Maybe it is vice-verse? It is not clear at all. Moreover, the common sense suggests that there is absolutely no conceptual difference between what are we waiting for: head or spindle positioning. Therefore, you twist the perfect sense into nonsense.
Actually, I prefer to think about latency as a "response time" rather just a delay. The head positioning better suits this definition than passive disk spinning. --Javalenok (talk) 09:36, 18 July 2011 (UTC)

The above discussion is preserved as an archive. Please do not modify it. Subsequent comments should be made in a new section.


Requested move[edit]

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review. No further edits should be made to this section.

The result of the move request was: No consensus to move. Sounds like the scope needs to be hammered out first. Cúchullain t/c 14:33, 30 November 2012 (UTC)



Hard disk drive performance characteristicsDisk drive performance characteristics – The recent addition of "hard" to the title, because "this is exclusively about HDDs", is not supported by the article text. Solid-state, optical and floppy disks are also part of the scope of the article. Relisted. Jenks24 (talk) 06:28, 4 November 2012 (UTC) Wbm1058 (talk) 17:50, 18 October 2012 (UTC)

Move log
I do not feel that restoring the hyphen is necessary. – Wbm1058 (talk) 17:57, 18 October 2012 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────

  • Strongly SupportThe lede clearly states the topic is broader than just HDDs by identifying drives with other types of media. Furthermore, much of the article is medium agnostic and in fact the performance characteristics of disk drives having different media is distinguished by degree not metric, so one article on disk drive performance characteristics makes sense. Rather than limit the article by title, we ought to be spending our time improving it. Tom94022 (talk) 22:56, 18 October 2012 (UTC)
  • 90% of the article presently covers HDDs exclusively. Occasionally the odd bit is thrown in about SSDs. Optical and floppy disk coverage is nonexistent. Our major problem in this entire sphere of articles is relentlessly creeping scope; far better to split out what little material relates to media other than HDDs to separate articles than to have to generalise content which was, perhaps inadvertently, written with an overwhelming focus on one particular medium. Chris Cunningham (user:thumperward) (talk) 11:43, 19 October 2012 (UTC)
I don't know where the 90% comes from but it doesn't appear that way to me. I really don't have time to go thru the article section by section but I would note that, for example, Section 1 is agnostic to device type. Perhaps a global search and replace HDD with disk drive were appropriate would make the article 90% agnostic. Tom94022 (talk) 17:06, 19 October 2012 (UTC)
Agree with Thumperward - leave it alone but if change is necessary I would further suggest initial capitalization for the compound word of Disk Drive is preferable. Tom94022 (talk) 17:06, 22 October 2012 (UTC)
Why? Disk drive isn't a proper noun or anything. --BDD (talk) 19:17, 25 October 2012 (UTC)
Tom94022's suggestion of caps, and Thumperward's reference to insertion of hyphen into the proper name High Street both display an ignorance of what's in WP:HYPHEN. Perhaps understanding that first would lead to a realization that hyphens used correctly will benefit the reader. Dicklyon (talk) 19:28, 25 October 2012 (UTC)
No personal attacks please, especially in the context of insisting that others read and adhere to Wikipedia policies! (;-> Andrewa (talk) 14:31, 26 October 2012 (UTC)
Unless the proper name is French, e.g., Trois-Rivières. Are we talking about:
  • hard-disk drive performance-characteristics (the performance characteristics of drives using hard disks),
  • hard disk-drive performance-characteristics (the performance characteristics of disk drives which are hard),
  • hard-disk drive-performance characteristics (the characteristics of drive performance of hard disks), or
  • hard-disk-drive performance-characteristics (the performance characteristics of hard disk drives)? Wbm1058 (talk) 19:32, 27 October 2012 (UTC)
Like I said, ignorance of WP:HYPHEN. Suggesting a hyphen in "performance-characteristics" is complete nonsense, not related to anything in our MOS or in any guide to English usage. The hyphens in "hard disk drive" when used attributively as here is widely supported, however, to help the reader parse the complicated combination of phrases, though some sources do say it's OK to omit the hyphens because "hard disk drive" may be treated as a permanent compound, found in the dictionary, as Tom94022 points out. Dicklyon (talk) 18:37, 31 October 2012 (UTC)
As Tom94022 helpfully points out below, hyphenation of compound words is regarded, in modern English at least, as a matter of style. When regarded as normative it frequently results in text which appears antiquated. I agree that capitalisation is not the correct move here, however. Chris Cunningham (user:thumperward) (talk) 14:52, 29 October 2012 (UTC)
Disk drive and hard disk drive are permanent compound phrases for which no ambiguity exists and therefore no hyphens are necessary, see Compound modifier, Exceptions especially note 1. To that I would add that the Authoritative Dictionary of IEEE Standard Terms (c) 2000 defines "disk drive" and "magnetic disk drive" without hyphens as does the Microsoft Computer Dictionary, 4th edition (c) 1999 and the latter includes "hard disk drive" without hyphens. Tom94022 (talk) 05:46, 28 October 2012 (UTC)

Disk drive performance characteristics – high level view[edit]

None of these links are specific, i.e., Hard disk access time, hard disk seek time, hard disk rotational latency. Are these characteristics limited to hard disk drives, or can other types of disk drives be measured by these characteristics? If so, do we need a separate link for floppy disk seek time, for example? – Wbm1058 (talk) 21:37, 13 November 2012 (UTC)

Access time, seek time, latency and data transfer rate are common characteristics of rotating media devices from Drums, thru HDDs to today's Blu-ray Disc optical disk drive (allowing for zero seek time in a drum). SSDs do not have rotating media but they appear to the system as an HDD and do have some of these characteristics. Treating them in one article makes sense to me. Tom94022 (talk) 01:17, 14 November 2012 (UTC)
The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page or in a move review. No further edits should be made to this section.

Data transfer rate - new terms[edit]

In the http://en.wikipedia.org/wiki/Hard_disk_drive_performance_characteristics#Data_transfer_rate, the terms "disk-to-buffer" and "buffer-to-computer" are introduced in the examples of throughputs. They are not discussed in the preamble or the definitions in this sections, and thus make the examples harder to understand. We could probably just put a: `aka "buffer-to-computer" ` in the preamble, where the different factors are described. — Preceding unsigned comment added by Myok (talkcontribs) 00:00, 31 December 2012 (UTC)

Relative importance of performance properties[edit]

This article does a good job of explaining what causes access times and data transfer rates. However, it does not indicate their relative importance. How much attention should one pay to transfer rate, versus the attention paid to access time? The answer depends on the workload considered. We could give a general answer based on studies, saying for example that server HDDs spend a third of their active time seeking and 2 thirds transferring, based on an average of many HDDs. Or, explain how to measure how a certain HDD's activity. --Chealer (talk) 16:26, 16 November 2013 (UTC)