Need prog. to prevent hdd going into idle (such as a perdiodic access?)

[Louis Grant]

Hello,

regarding my laptop hardddrive Fujitsu MHV2060AT - there seems to be no
way to change the APM settings, and these are set at going into active
idle after a few seconds and a few seconds thereafter going into idle
with parked heads. The result is that when using Word for example, the
hdd clicks and clicks because the hdd keeps changing states. According
to Fujitsu there is nothing I can do.

a) Is Fujitsu telling the truth?

b) What about a programm that reads any odd bit from my hdd at 1 second
intervals? or sends some sort of wakeup command? so that the hdd doesnt
get the chance to go into idle....

thanks,
Louis

 

[Arno Wagner]

Hello,

regarding my laptop hardddrive Fujitsu MHV2060AT - there seems to be no
way to change the APM settings, and these are set at going into active
idle after a few seconds and a few seconds thereafter going into idle
with parked heads. The result is that when using Word for example, the
hdd clicks and clicks because the hdd keeps changing states. According
to Fujitsu there is nothing I can do.

a) Is Fujitsu telling the truth?

Not likely. There is nothing Fujitsu can do, since APM is not a function
of the disk, but rather the OS. Unless you are talking about the auto
spindown of the drive? That can be changed. Don't know the right
software for Windows. Mybe the Hitachi HDD feature tool can help.
Maybe Sisoft Sandra can do it.

Under Linux "hdparm -S <value> <drive>" does
it. The encoding is a bit strange, so you need to read the man-page.
And if you are talking about the disk APM, not the OS apm, that

b) What about a programm that reads any odd bit from my hdd at 1 second
intervals? or sends some sort of wakeup command? so that the hdd doesnt
get the chance to go into idle....

Why not change the APM settings to send the disk to sleep later?
The disk access every second would cause quite a slowdown
and uneccesary power drain.

Arno

 

[Eric Gisin]

What does Power Option Settings say?

 

[Louis Grant]

Not likely. There is nothing Fujitsu can do, since APM is not a function
of the disk, but rather the OS. Unless you are talking about the auto
spindown of the drive? That can be changed. Don't know the right
software for Windows. Mybe the Hitachi HDD feature tool can help.
Maybe Sisoft Sandra can do it.

This is what it says:
--------------
Table 1.7 Advanced Power Management
APM Mode
Active Idle (VCM Lock) Low Power Idle (Unload) Standby (Spin Off)
Mode-0 0.2-1.2 sec 15 min. N/A
Mode-1 0.2-1.2 sec 10.0-40.0 sec N/A
Mode-2 0.2-1.2 sec 10.0-40.0 sec 10.0-40.0 sec
When the maximum time that the HDD is waiting for commands has been
exceeded:
Mode-0: Mode shifts from Active condition to Active Idle in 0.2-1.2,
and to Low Power Idle in 15 minutes.
Mode-1: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds.
Mode-2: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds. After 10.0-40.0
seconds in Low Power Idle, the mode shifts to standby.
------------

The drive is shipped in Mode 1. There are no other modes to the above,
i.e. active idle is always on. Apart from that there is no way to
change modes in xp. Hitachi Tool wont do it either.

You can probably tell from the "0.2-1.2 sec" how annoying this is. on,
off, on, off, etc. As the drive is built in closest to me and not very
soundproof. I can hear it clearly.

Under Linux "hdparm -S <value> <drive>" does
it. The encoding is a bit strange, so you need to read the man-page.
And if you are talking about the disk APM, not the OS apm, that
cab be disabled with "hdparm -B 255 <device>".

Using Windows XP. Total user, i.e. no linux oder prog. skills.

Why not change the APM settings to send the disk to sleep later?

It doesnt go to sleep. See above, it has an active idle and a low power
idle where it unloads head after 7-10 secs as far as i can tell.

The disk access every second would cause quite a slowdown
and uneccesary power drain.

it only needs to read one bit. when I use word, I dont mind a slight
slowdown. power is no problem, as I work on plug most of the time.

Ingolf

 

[Louis Grant]

What does Power Option Settings say?

it's not that, that's all "off" for hdd.

Ingolf

 

[Louis Grant]

Louis Grant schrieb:

Ingolf

getting confused with my names, it's Louis of course...

Louis a.k.a. Ingolf

 

[Rod Speed]

Arno Wagner wrote

Why ? Word shouldnt be using the drive that often.

And see below.

I bet you asked them the wrong question.

This is what it says:
--------------
?Table 1.7 Advanced Power Management
APM Mode
Active Idle (VCM Lock) Low Power Idle (Unload) Standby (Spin Off)
Mode-0 0.2-1.2 sec 15 min. N/A
Mode-1 0.2-1.2 sec 10.0-40.0 sec N/A
Mode-2 0.2-1.2 sec 10.0-40.0 sec 10.0-40.0 sec
When the maximum time that the HDD is waiting for commands has been
exceeded:
Mode-0: Mode shifts from Active condition to Active Idle in 0.2-1.2,
and to Low Power Idle in 15 minutes.
Mode-1: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds.
Mode-2: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds. After 10.0-40.0
seconds in Low Power Idle, the mode shifts to standby.
------------

The drive is shipped in Mode 1. There are no other
modes to the above, i.e. active idle is always on.

What is wrong with Mode 0 ? I bet you cant hear it going
from Active to Active Idle, what you are hearing is it going
to Low Power Idle. So Mode 0 should fix what you dont like.

Apart from that there is no way to change
modes in xp. Hitachi Tool wont do it either.

You can probably tell from the "0.2-1.2 sec" how annoying
this is. on, off, on, off, etc. As the drive is built in closest
to me and not very soundproof. I can hear it clearly.

Word shouldnt be using the drive all that often, and changing the
shift from Active Idle to Low Power Idle to 15 mins should stop
you hearing anything, because it will stay out of Low Power Idle.

Using Windows XP. Total user, i.e. no linux oder prog. skills.

It doesnt go to sleep. See above, it has an active idle and a low
power idle where it unloads head after 7-10 secs as far as i can tell.

Yes, and that will be what you are hearing.

You wont be able to hear it changing from Active to Active Idle.

it only needs to read one bit. when I use word, I dont mind a slight
slowdown. power is no problem, as I work on plug most of the time.

Makes more sense to change to Mode 0.

 

[Arno Wagner]

Arno Wagner schrieb:

This is what it says:
--------------
???Table 1.7 Advanced Power Management
APM Mode
Active Idle (VCM Lock) Low Power Idle (Unload) Standby (Spin Off)
Mode-0 0.2-1.2 sec 15 min. N/A
Mode-1 0.2-1.2 sec 10.0-40.0 sec N/A
Mode-2 0.2-1.2 sec 10.0-40.0 sec 10.0-40.0 sec
When the maximum time that the HDD is waiting for commands has been
exceeded:
Mode-0: Mode shifts from Active condition to Active Idle in 0.2-1.2,
and to Low Power Idle in 15 minutes.
Mode-1: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds.
Mode-2: Mode shifts from Active condition to Active Idle in 0.2-1.2
seconds and to Low Power Idle in 10.0-40.0 seconds. After 10.0-40.0
seconds in Low Power Idle, the mode shifts to standby.
------------

The drive is shipped in Mode 1. There are no other modes to the above,
i.e. active idle is always on. Apart from that there is no way to
change modes in xp. Hitachi Tool wont do it either.

You can probably tell from the "0.2-1.2 sec" how annoying this is. on,
off, on, off, etc. As the drive is built in closest to me and not very
soundproof. I can hear it clearly.

I see. I have a bit older Fujitsu HDD in my laptop (I think
it is an MHT) and it does not go into power-save mode at all,
at least I never noticed so far.

Still, shifting this to Mode-0 should provide some improvement.
Maybe it has acousic management? If you turn that to "silent",
the active-idle might become inaudible.

Using Windows XP. Total user, i.e. no linux oder prog. skills.

Maybe somebody else here knows how to do that. I use XP only
for gaming, I would not know how to do anything like this with it.

It doesnt go to sleep. See above, it has an active idle and a low power
idle where it unloads head after 7-10 secs as far as i can tell.

it only needs to read one bit. when I use word, I dont mind a slight
slowdown. power is no problem, as I work on plug most of the time.

It will cause a seek. Hmmm. I still don't think this will work
well. Have a look whether there is acoustic management first.

Arno

 

[Louis Grant]

Why ? Word shouldnt be using the drive that often.

But Windows does, plus I perdiodically save my documents.


[...]

Yes, and that will be what you are hearing.

Wrong. See below.

You wont be able to hear it changing from Active to Active Idle.

Yes I do, active idle and low power idle, two different actions /
noises at different times. I admit that it also has to do with the
laptop, an ibm R52 that has the hdd built in front, as opposed to an
A30, which had it built in back bottom. Yet even in the latter I could
hear it, though it was not annoying as it is with the R52.

Makes more sense to change to Mode 0.

Unfortunately not, see above.

Louis

 

[Louis Grant]

Arno Wagner schrieb:

Still, shifting this to Mode-0 should provide some improvement.
Maybe it has acousic management? If you turn that to "silent",
the active-idle might become inaudible.

See reply to "Rod Speed".

It will cause a seek. Hmmm. I still don't think this will work
well. Have a look whether there is acoustic management first.

Accoustic management is set to quiet. Trust me, I've been though all
this The drive as such is really quiet when accessing or in any of
the modes, quieter than any other drive I know. Just when it switches
states it's a pain in the neck, whereas other drives dont make a noise
at all.

Louis

 

[Rod Speed]

Rod Speed wrote

But Windows does,

Nope, not if you set it up properly.

plus I perdiodically save my documents.

Mode 0 will be fine with that as long as you save
them more often than every 15 mins. I certainly do.

Nope.

See below.

Didnt help.

Yes I do, active idle and low power idle, two
different actions / noises at different times.

Dont believe it. I bet what you are actually hearing is two
different sounds, entering and leaving Low Power Idle.

I admit that it also has to do with the laptop, an ibm R52
that has the hdd built in front, as opposed to an A30, which
had it built in back bottom. Yet even in the latter I could
hear it, though it was not annoying as it is with the R52.

See above.

Unfortunately not, see above.

You havent establised that you can hear it change from Active to Active Idle.

 

[Fabien LE LEZ]

b) What about a programm that reads any odd bit from my hdd at 1 second
intervals?

That ain't so easy to do.
OTOH, a program that reads a random byte from a big file every n
milliseconds is far easier.

I'm not sure it'll be helpful to you, but here goes:
<http://perso.edulang.com/csiphs/access_hard_drive.zip>

To use it, you have to put the .exe somewhere on your hard disk, then
create a shortcut with some command-line arguments.
New -> Shortcut ->
(path)\access_hard_drive.exe a_very_big_file 1000
The last argument is the "sleep time" between two accesses, in
milliseconds. 1000 = 1 second; you can put whatever you want there.

The .cpp is the source code; I don't think it'll be of any use to you
-- if you could read it, you'd already have done roughly that same
program.

If you don't have a very big file, you can download a 1500 MB file
(compressed to 1.45 MB) here:
<http://perso.edulang.com/csiphs/access_hard_drive.dummy.zip>

If you use that file, and you put both it and the .exe in
c:\antisleep, and you want an access every 0.5 seconds, the
command-line of the shortcut would be:
c:\antisleep\access_hard_drive.exe c:\antisleep\dummy 500

 

[Louis Grant]

That ain't so easy to do.
not?

OTOH, a program that reads a random byte from a big file every n
milliseconds is far easier.

I'm not sure it'll be helpful to you, but here goes:
<http://perso.edulang.com/csiphs/access_hard_drive.zip>

[...]

thanks! did you write that?

only does the trick for a little while though (20-30 secs?), then it
seems that Windows (?) reads the file into memory, i.e. no regular
harddrive access. The numbers still keep appearing regularly, but the
hdd-access isnt... ... i.e. it's not that easy then after all? got any
other ideas?

thanks again!

Louis

 

[Arno Wagner]

Fabien LE LEZ schrieb:

That ain't so easy to do.

not?

OTOH, a program that reads a random byte from a big file every n
milliseconds is far easier.

I'm not sure it'll be helpful to you, but here goes:
<http://perso.edulang.com/csiphs/access_hard_drive.zip>

[...]

thanks! did you write that?

only does the trick for a little while though (20-30 secs?), then it
seems that Windows (?) reads the file into memory, i.e. no regular
harddrive access. The numbers still keep appearing regularly, but the
hdd-access isnt... ... i.e. it's not that easy then after all? got any
other ideas?

The only reliable thing against caching is to read from a file
larger than your main memory.

Arno

 

[Louis Grant]

Arno Wagner schrieb:

[...]

The only reliable thing against caching is to read from a file
larger than your main memory.

Sounds plausibel.... yet, the dummy file is 1,5 Gb, my RAM is 512 Mb.

Could it be that XP reads the dummy file into memory and apply some
data compression? in which case I would need to use a compressed dummy
file larger than my main memory?

Louis

 

[Rod Speed]

Arno Wagner wrote

Its not quite that bad, XP doesnt use all of main memory for a cache.

Sounds plausibel.... yet, the dummy file is 1,5 Gb, my RAM is 512 Mb.

Could it be that XP reads the dummy file into
memory and apply some data compression?
Nope.

in which case I would need to use a compressed
dummy file larger than my main memory?

Nope, wont help, try it.

It cant be reading the whole of that 1.5G file in that time.

 

[Arno Wagner]

Arno Wagner schrieb:

[...]

The only reliable thing against caching is to read from a file
larger than your main memory.

Sounds plausibel.... yet, the dummy file is 1,5 Gb, my RAM is 512 Mb.

Could it be that XP reads the dummy file into memory and apply some
data compression?

No. While a theroretical posibility, that is not done today.

in which case I would need to use a compressed dummy
file larger than my main memory?

What is possible is that the reader program does not distribute its
reads evenly over the file. If it only reads from a, say, 100MB
selection, the XP would buffer that.

Arno

 

[Fabien LE LEZ]

thanks! did you write that?

Yep. As you can see in the source code, it's a really trivial program.

only does the trick for a little while though (20-30 secs?), then it
seems that Windows (?) reads the file into memory, i.e. no regular
harddrive access.

Weird. There's not way Windows can store 1.5 GB of data in its RAM
cache.

 

[Fabien LE LEZ]

What is possible is that the reader program does not distribute its
reads evenly over the file.

That's not the case. I just made a test run, and got the following
results (in the format "offset nb_bytes_read"):
[Note: there's no 32-bit-related problems since the file is less than
2 GB.]

16608190 1 527137596 1 559498425 1 844569397 1 1101549596 1
432213325 1 171441515 1 887617992 1 259815143 1 1078260691 1
1301758057 1 345140486 1 1498312297 1 1452134923 1
709421028 1 1040716786 1 863794561 1 179251497 1 191024220 1
746951572 1 887488535 1 1366380691 1 282836307 1 1092775857 1
846702983 1 980496146 1 1238172908 1 725202554 1 592683894 1
949165431 1 268310648 1 463984027 1 642731540 1 651122606 1 777167217
1 2563671 1 109683547 1 400030784 1 558547436 1 310089460 1
332939953 1 1192458689 1 1424051325 1 725593515 1 55494268 1
720407231 1 938940154 1 180273664 1 1021242540 1 1377122052 1
870600514 1 976501013 1 398986416 1 371200103 1 890598376 1 966049491
1 1492138210 1 1271262197 1 1124723155 1 868155437
1 264207954 1 1367655573 1 966909771 1 1559279140 1
1160088469 1 464626465 1 449550712 1 774488408 1 879522428 1
1218162656 1 1023702099 1 227920143 1 1445455414 1
1019986048 1 1031071747 1 1134047888 1 662378986 1
592559214 1 23085221 1 711419390 1 859367738 1 985089973 1 917461908
1 680256177 1 1314322367 1 1551957946 1 555021080 1
1013691107 1 1098682727 1 487339120 1 113565946 1 963795930
1 1332610902 1 1061276405 1 776605721 1 1070894862 1
557940258 1 1537388205 1 384643696 1 1436900137 1 532554518
1 229350977 1 507031286 1 982077574 1 423892806 1 1540763793 1
340948098 1 830154136 1 1423500734 1 967098902 1 1030660002 1
812487496 1 1333715440 1 966026291 1 1532687179 1 101009602
1 1148068664 1 1071664254 1 182130195 1 396936233 1
32015816 1 1923947 1 388826057 1 665050564 1 722743018 1 321907957
1 13846282 1 716072732 1 184868190 1 1507010857 1 603936950
1 1052907390 1 122404094 1 281385778 1 307980776 1 554209280
1 1172887410 1 1375488960 1 358265167 1 1432786789 1
1229897791 1 956222326 1 1525708285 1 1020987512 1
1040222921 1 923674035 1 1558083109 1 267865817 1 1347365592
1 598425054 1 1464791505 1 636436403 1 1016896756 1
401853711 1 202636275 1 411205873 1 864646585 1 227790263 1 1354303232
1 1425742278 1 603141545 1 370561454 1 342609254 1 1044842699
1 1489391446 1 289729088 1 939588352 1 858194796 1 519018511
1 1504130578 1 1301007729 1 628876380 1 1331736438 1
458629520 1 1431401847 1 1088164968 1 934483468 1 1248232241
1 622056481 1 1423473351 1 999649432 1 1013938382 1
590582791 1 893571148 1 1273180831 1 1487317442 1 166257380
1 1427956381 1 256083884 1 1070803351 1 395784353 1
198409030 1 15565604 1 484187026 1 1194871388 1 475670592 1
968404845 1 1529402940 1 555571795 1 1448641652 1 793865383
1 547433681 1 1028080570 1 37201441 1 620246960 1 159332511
1 1457320473 1 1551577399 1 1185298360 1 1543299937
1 712142074 1 1451208737 1 1551916467 1 840486266 1
824993835 1 1116128000 1 380043323 1 699376172 1 104542151 1
1008956627 1 1327049198 1 783459253 1 1386921426 1
1110682520 1 557225649 1 271946603 1 823038137 1 767771603 1
399716580 1 1343005364 1 1241832885 1 1093653380 1
396919311 1 1058087296 1 1330428467 1 490372503 1 345547634
1 1529847333 1 1526683940 1 589558335 1 1175252879 1
626308154 1 1456602672 1 593768506 1 518882972 1 959862738 1
1501522417 1 465438487 1 275883510 1 984186931 1 628943371 1
379805138 1 1002344803 1 1230550004 1 1498338234 1
1161308283 1 430926960 1 517171219 1 821464278 1 984206263 1
836998929 1 1005462392 1 1152089405 1 920653510 1 87087577 1
83972457 1 66956205 1 461632096 1 971588744 1 536546017 1 776256032
1 312577305 1 1073292303 1 1246631926 1 124477053 1
1453217191 1 920361234 1 1467261831 1 1166970319 1
462106142 1 539828675 1 1355482750 1 1564217821 1 1552748447
1 139196988 1 344539630 1 260074555 1 1280975846 1 198890820
1 1216879903 1 425136594 1 974443105 1 974556174 1 1005292057
1 799331133 1 1270561574 1 1409445463 1 177970193 1
601376563 1 783446753 1 602952569 1 223904595 1 702120529 1 471783843
1 261953918 1 648646280 1 1062620002 1 1126427320 1
271067534 1 98509573 1 727651321 1 267510498 1 1537462100 1
40828663 1 376042280 1 1034889082 1 381048027 1 1562940164 1
1134306174 1 41198739 1 1209746509 1 8627687 1 1228505869
1 1374310657 1 293340598 1 1044412386 1 1483099855 1
125217105 1 404260246 1 1533497539 1 792546321 1 1501827003 1
1276357392 1 1276935096 1 441041215 1 31919478 1 483047004
1 9006251 1 1539974295 1 642551427 1 17998151 1 94671998 1
303423205 1 1402351086 1 1275095064 1 772645698 1 35045534 1
838425578 1 1012164204 1 1245543641 1 505152871 1 1518586497
1 879939309 1 35065601 1 893790879 1 460557451 1 633432478 1
38429413 1 570609348 1 971219631 1 189663212 1 412027523 1 858642264
1 303703378 1 1096385493 1 448465552 1 1489149868 1
316949742 1 1119156840 1 1290630867 1 1184898718 1
244238042 1 821501366 1 764583137 1 800407753 1 1454999433 1
118262851 1 205538888 1 1152536360 1 616188466 1 530259799 1
1256950086 1 262070426 1 1138972651 1 558050502 1 506708679
1 1015250408 1 441828020 1 1475688098 1 835085170 1
934217397 1 217448685 1 1524321641 1 152801177 1 231289171 1
380814411 1 526891157 1 1569379995 1 1410154270 1 291712167
1 766316709 1 951583164 1 412112050 1 960687696 1 982845462 1
630525018 1 765694747 1 373291076 1 791772740 1 499113265 1 407687542
1 807464276 1 1365377096 1 1102679534 1 1507770622 1
1405958731 1 806349369 1 1420527949 1 828111239 1 1167638892
1 1514411739 1 777121686 1 1514049780 1 1350599798 1
1234910811 1 162875531 1

 

[Louis Grant]

Fabien LE LEZ schrieb:

[...]

Weird. There's not way Windows can store 1.5 GB of data in its RAM
cache.

true. I wonder how xp manages to run the prog. for long intervals
without hdd usage... if it's not caching? damn windows...

Louis