Message boards : Number crunching : constant HD activity?
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Profile scsimodo

Send message
Joined: 17 Sep 04
Posts: 12
Credit: 54,554
RAC: 0
Message 7266 - Posted: 28 Apr 2005, 9:17:59 UTC

Anyone else experiencing a constant (every second) HD activity when running those new LHC units?

On my PC (Athlon XP 2600+, Win2k), LHC seems to write to disk every(!) second (on other PCs too). It's quite annoying. All other other projects (S@H, E@H, P@H) run "quiet", an update is made at most every minute.

Older LHC WUs (a few weeks ago) didn't show this behaviour)

ID: 7266 · Report as offensive     Reply Quote
Profile Seventh Serenity
Avatar

Send message
Joined: 27 Sep 04
Posts: 51
Credit: 72,804
RAC: 0
Message 7267 - Posted: 28 Apr 2005, 9:35:25 UTC

The drive I have LHC@Home running off (Seagate Medalist 3GB 5200RPM 256KB cache) does the same thing you describe. It doesn't bother me, as the drive isn't used by anything else (so LHC@Home has the whole drive) and it lets me know that it's running when I'm doing something else.

Is it possibly a checkpoint on the new WU's?
ID: 7267 · Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 2 Sep 04
Posts: 545
Credit: 148,912
RAC: 0
Message 7273 - Posted: 28 Apr 2005, 10:40:35 UTC

Hmmm,

I don't see this. I jsut see the normal "flash" once a second ...

Is this an older drive? Could it be going bad? Do you have other programs running?

Remember, by default Microsoft wants to index disk drives, even though that does not seem to speed up access (in my experience), and with constant changes to the drive this may drive the indexing service crazy.


ID: 7273 · Report as offensive     Reply Quote
Petu

Send message
Joined: 17 Sep 04
Posts: 1
Credit: 16,031
RAC: 0
Message 7278 - Posted: 28 Apr 2005, 12:02:37 UTC

I have noticed the same thing. I can tell by the sound of my HD (a fairly new Seagate SATA-drive) when LHC is running...crunch,crunch,crunch... Other projects run silently.
ID: 7278 · Report as offensive     Reply Quote
STE\/E

Send message
Joined: 2 Sep 04
Posts: 352
Credit: 1,393,150
RAC: 0
Message 7280 - Posted: 28 Apr 2005, 12:10:57 UTC

can tell by the sound of my HD (a fairly new Seagate SATA-drive) when LHC is running...crunch,crunch,crunch...
=========

Thats just the sound of a happy computer ... hehe ... ;)
ID: 7280 · Report as offensive     Reply Quote
Profile littleBouncer
Avatar

Send message
Joined: 23 Oct 04
Posts: 358
Credit: 1,439,205
RAC: 0
Message 7283 - Posted: 28 Apr 2005, 12:20:36 UTC
Last modified: 28 Apr 2005, 18:12:43 UTC

Since(one and half month ago) the HD of my Notebook crashed, I let run the client on an external HD-drive(Western Digital). For that I noticed:

When "show graphic" is active then the HD is in permanent use.
In 'still processing-mode without graphic': there is every sec. (not how it is set in the prefs.) access to the HD.

greetz littleBouncer
ID: 7283 · Report as offensive     Reply Quote
PeteHow

Send message
Joined: 1 Sep 04
Posts: 3
Credit: 28,554
RAC: 0
Message 7294 - Posted: 28 Apr 2005, 14:33:44 UTC

I commented on this disk activity 2 months ago: http://lhcathome.cern.ch/forum_thread.php?id=1160

Markku Degerholm suggested that since Sixtrack predates BOINC, it does not recognise any BOINC settings. I still find the constant Tick Tick Tick... of the hard drive a little annoying, and wonder if it is adversely affecting the disk's lifetime. Earlier WUs (or is it client?) did not exhibit this characteristic.

ID: 7294 · Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 2 Sep 04
Posts: 545
Credit: 148,912
RAC: 0
Message 7298 - Posted: 28 Apr 2005, 15:50:20 UTC - in response to Message 7294.  

> I commented on this disk activity 2 months ago:
> http://lhcathome.cern.ch/forum_thread.php?id=1160
>
> Markku Degerholm suggested that since Sixtrack predates BOINC, it does not
> recognise any BOINC settings. I still find the constant Tick Tick Tick... of
> the hard drive a little annoying, and wonder if it is adversely affecting the
> disk's lifetime. Earlier WUs (or is it client?) did not exhibit this
> characteristic.

If the drive is up and spinning that is one of the most common failure modes, the bearing die. But, I run my computers 24x7x365 and the drives are never sleep ... I have never had one die on me yet ...

ID: 7298 · Report as offensive     Reply Quote
Comet_SL9

Send message
Joined: 18 Sep 04
Posts: 1
Credit: 31,192
RAC: 0
Message 7299 - Posted: 28 Apr 2005, 17:25:01 UTC

I have the same thing here; Athlon XP 2600+ with RAID 0 HD System (~70 GB). The HD is flashing every second although the BOINC setting says "Write to disk at most every 60 seconds"...

It seems to be a real bug, LHC@home ignores BOINC settings...

ID: 7299 · Report as offensive     Reply Quote
Profile adrianxw

Send message
Joined: 29 Sep 04
Posts: 187
Credit: 705,487
RAC: 0
Message 7300 - Posted: 28 Apr 2005, 18:04:34 UTC

I see this as well. 3 month old SATA Seagate Barracuda with 8MB Cache - flash flash flash. Suspend project, (all other projects are suspended at the moment), disk returns to the once a second, resume project, constant activity.

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.
ID: 7300 · Report as offensive     Reply Quote
Travis DJ

Send message
Joined: 29 Sep 04
Posts: 196
Credit: 207,040
RAC: 0
Message 7302 - Posted: 28 Apr 2005, 21:34:55 UTC

Same behavior here. Preferences are set to write to disk at most 300 seconds (5 minute intervals). Sixtrack is not honoring the preferences for some unknown reason.

My Hardware:
Biostar M7NCD nForce2 250
AMD AthlonXP 3200+ (not OCed)
Corsair XMSPro DDR400 2-2-2-5-1T
Seagate 7200.6 120GB PATA HDD
WinXP Pro SP2/UpToDate
etc..

ID: 7302 · Report as offensive     Reply Quote
Profile Bruno G. Olsen & ESEA @ greenh...
Avatar

Send message
Joined: 17 Sep 04
Posts: 52
Credit: 247,983
RAC: 0
Message 7305 - Posted: 28 Apr 2005, 22:44:16 UTC

I'm not seeing this (boinc 4.32, sixtrack 4.67, MQonly-wu's)


ID: 7305 · Report as offensive     Reply Quote
grumpy

Send message
Joined: 1 Sep 04
Posts: 57
Credit: 2,835,005
RAC: 0
Message 7308 - Posted: 29 Apr 2005, 1:42:00 UTC
Last modified: 29 Apr 2005, 1:43:57 UTC

it's writing about every seconds to the fort.** files
I don't get activity like that with seti/einstein



6 21:35:04 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.90 SUCCESS Offset: 106840 Length: 148

7 21:35:04 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.88 SUCCESS Offset: 106840 Length: 148

8 21:35:04 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.85 SUCCESS Offset: 106840 Length: 148

9 21:35:05 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.91 SUCCESS Offset: 0 Length: 15

10 21:35:05 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.90 SUCCESS Offset: 106988 Length: 148

11 21:35:05 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.88 SUCCESS Offset: 106988 Length: 148

12 21:35:05 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.85 SUCCESS Offset: 106988 Length: 148

13 21:35:06 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.91 SUCCESS Offset: 0 Length: 15

14 21:35:06 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.90 SUCCESS Offset: 107136 Length: 148

15 21:35:06 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.88 SUCCESS Offset: 107136 Length: 148

16 21:35:06 sixtrack_4.67_w:424 WRITE C:Program FilesBOINCslots1fort.85 SUCCESS Offset: 107136 Length: 148



ID: 7308 · Report as offensive     Reply Quote
Profile Juerschi

Send message
Joined: 26 Oct 04
Posts: 12
Credit: 8,909
RAC: 0
Message 7324 - Posted: 29 Apr 2005, 18:12:22 UTC

No I have noticed this behavior too.

Runing sixtrack 4.67 on XP SP2 with MQonly WU's
Preferences set to write every 300 seconds


ID: 7324 · Report as offensive     Reply Quote
Profile ksba

Send message
Joined: 27 Sep 04
Posts: 40
Credit: 1,742,415
RAC: 0
Message 7327 - Posted: 29 Apr 2005, 20:48:53 UTC - in response to Message 7324.  

> No I have noticed this behavior too.

Jep, the new WU thake down my raid5 :-)

If 30 fast p4 run all on the same (gigabitswitched) network share (at night) than goes the network ;-) and cpu/raid processor some happy percents higher ...
At the "MD5 - WU error" my raid5 was near of overheating.. and the servercpu goes up to the limit. lhc is a great test for the new raid processor ;-)

ID: 7327 · Report as offensive     Reply Quote
ca_grufti

Send message
Joined: 23 Oct 04
Posts: 2
Credit: 117,212
RAC: 0
Message 7369 - Posted: 1 May 2005, 1:04:49 UTC

Non stop. It needs to get fixed. It's one thing to donate cycles, it's totally another to risk HD failure.
ID: 7369 · Report as offensive     Reply Quote
grumpy

Send message
Joined: 1 Sep 04
Posts: 57
Credit: 2,835,005
RAC: 0
Message 7377 - Posted: 1 May 2005, 16:06:33 UTC

HD fragmentation will exacerbate this "problem" and make it noisier.

I don't like the idea of having my HD actuator motor
behaving like that.
ID: 7377 · Report as offensive     Reply Quote
Profile littleBouncer
Avatar

Send message
Joined: 23 Oct 04
Posts: 358
Credit: 1,439,205
RAC: 0
Message 7383 - Posted: 1 May 2005, 22:18:02 UTC

Try to let the client run from a MP3-Player, used as a USB-Drive!

greetz littleBouncer
ID: 7383 · Report as offensive     Reply Quote
Aurora Borealis

Send message
Joined: 18 Sep 04
Posts: 59
Credit: 317,857
RAC: 0
Message 7387 - Posted: 2 May 2005, 4:32:59 UTC - in response to Message 7383.  

> Try to let the client run from a MP3-Player, used as a USB-Drive!
>
> greetz littleBouncer
>

A warning.

I believe, I read somewhere, that USB-Drive, and I presume any MP3-Players also, have a finite number of erase/write cycles. Continuous usage in this way could have a negative effect on its life expectancy.

Can anyone confirm or refute this.

Questions? Answers are in the BOINC Wiki.

Boinc V6.10.56 Recommended
WinXP C2D 2.1G 3GB
ID: 7387 · Report as offensive     Reply Quote
Profile Alex

Send message
Joined: 2 Sep 04
Posts: 378
Credit: 10,765
RAC: 0
Message 7388 - Posted: 2 May 2005, 4:35:56 UTC - in response to Message 7387.  
Last modified: 2 May 2005, 4:38:26 UTC

>
> Can anyone confirm or refute this.
>
>

I've heard this, but cannot confirm or refute this.

I think the main cause of USB drives 'failing' is that the FAT table gets fouled up when unplugging from a host

I also ran Boinc on my usb drive for a while on my old laptop, and moved it from pc to pc. So I was able to confirm that moving from a pentium1 to a pentium 2 to a AMD k6 to a 2.2G celeron was able to have a validated work unit.

I actually doubt the read/write cycle thing.
Samsung is coming out with a hybrid flash/hard disk drive to have lower energy drives.
I'm not the LHC Alex. Just a number cruncher like everyone else here.
ID: 7388 · Report as offensive     Reply Quote
1 · 2 · Next

Message boards : Number crunching : constant HD activity?


©2024 CERN