|
I suggest you calculate the number of bytes moved in between two hiccups, and I bet it will be a multiple of the sector size (512B), possibly simply the cluster size (which could be the next power of 2 multiple of 512B that exceeds the partition size divided by 64K).
What probably is happening is a new cluster needs to be allocated, causing an update of the FAT table. A possible workaround would be to preallocate the file, which implies you need to know and allocate the maximum size before you start writing the data. A possible approach is by using a "memory mapped file".
AFAIK solid state disks are faster, especially when reading data and hopping around (their seek time is much better). So the same problem would exist, at a much smaller scale.
Luc Pattyn [Forum Guidelines] [My Articles] Nil Volentibus Arduum
Please use <PRE> tags for code snippets, they preserve indentation, improve readability, and make me actually look at the code. modified on Tuesday, January 11, 2011 7:41 PM
|
|
|
|
|
Luc Pattyn wrote: AFAIK solid state disks are faster, especially when reading data and hopping around (their seek time is much better). So the same problem would exist, at a much smaller scale.
You have a potential much bigger problem with SSDs.
You cant erase single cells in an SSD, but instead it erases pretty large blocks of data (usually 128kB).
So if you change a few bytes in a file, the drive will write another block, and mark the original block for deletion, and do the actual delete whenever it's idle.
If you use an SSD for logging purposes, it might never get idle for long enough to get the erasing done and you would end up with a drive that's not having any free blocks despite being far from full.
When this happens you will get a hiccup of a totally different magnitude.
|
|
|
|
|
Interesting. So for continuous writes, the app should not flush, and one should have Windows cache the file in chunks that correspond to the block size of the SSD, avoiding almost all partial-block writes. Not sure that can be organized easily.
Luc Pattyn [Forum Guidelines] [My Articles] Nil Volentibus Arduum
Please use <PRE> tags for code snippets, they preserve indentation, improve readability, and make me actually look at the code.
|
|
|
|
|
Even in a degraded state (and forcing that is getting hard even for benchmarkers) a current generation SSD will still outperform a mechanical drive.
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
The original response talks about preallocating files. I'd like to try an experiment with this approach and just see what the improvement is. Can anyone suggest some VS c++ code that will accomplish this. 1) preallocated file creation, that would be reused with each running of the program 2) code that opens the file 3) code that sequencially writes ascii data msgs to the file starting at the beginning.
Thanks
|
|
|
|
|
Yes it will indeed.
And in all normal use you won't notice it at all.
|
|
|
|
|
The original response talks about preallocating files. I'd like to try an experiment with this approach and just see what the improvement is. Can anyone suggest some VS c++ code that will accomplish this. 1) preallocated file creation, that would be reused with each running of the program 2) code that opens the file 3) code that sequencially writes ascii data msgs to the file starting at the beginning.
Thanks
|
|
|
|
|
Hi guys!
I'm concerning situation, what will do the job quicker. One server hdd, for example WD VelociRaptor, 3.5'', 300GB, Serial ATA/300, 10000RPM, 16MB cache, or two WD Caviar Black, 3.5'', 1TB, SATA/600, 7200RPM, 64MB cache connected into RAID 0.
For me, intuitively raid will be quicker, but some test or personal experience will be useful. Unfortunately I never worked with raid so I only know the theory about it.
I know that there is double chance to data loss with raid 0 but I'm planning to buy third external hdd which will be working as a backup storage.
I don't wan't to create raid 1+0 because of lack of founds to buy more hdd's.
In the price of one 300GB server disk I can get two 1 TB disks.
What are you thinking about such situation? The computer will work as a render machine, mostly for 3ds max and autocad as well, so the speed of disk is quite important matter for me.
Oh! And I'm planning to use raid controller to relieve processor from counting all that stripe stuff.
|
|
|
|
|
First of all, there isn't as big difference performancewise between those disks as you might think.
The velociraptor has only a slightly higher sustained datarate, and if you would consider the 1.5 TB Caviar, it's even slower.
But the latency is 30% lower.
It's also serverclass when it comes to reliability, which the caviar certainly isn't.
So it's more about how you're going to use the drives that makes the difference.
If you read or write very large contiguous files in one go, the raid system is the clear winner.
But if you read or write very many small files, then the single velociraptor will be the better choice.
Also keep in mind that a cheap raid controller isn't fast. Check the sustained write speed if available.
If you really want a fast system, you should consider one of these[^] instead.
|
|
|
|
|
Does anyone know where I might find application benchmarks for dual socket core i5 2500K versus one and two socket core i7 2600K? Curious about Excel and compiler benchmarks.
Is overclocking effective with dual socket machines with or without liquid cooling? Note that this will be for my personal machine that I will program on at home and not for a server, etc.
|
|
|
|
|
I haven't seen any 2 socket benchmarks yet. Annand did a compiler benchmark with single socket systems.
The catch with overclocking is that in general 2 socket boards are intended for servers and have minimal hardware support (most notably they typically don't have VRMs capable of handling the excess load, or the extra heat it produces) even if it's allowed in the BIOS.
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
Thanks. I'll probably go with single socket overclocking 2600K or dual socket 2500 without overclocking if I find a two socket board in the next month.
|
|
|
|
|
I didn't occur to me at the time, but you're not going to find a 2 socket 1155 board because it's not a supported configuration (no QPI to link the chips). On the Intel side you'll have to wait until Q3 for LGA 2011 to launch before you can get multi-socket support for Sandybridge.
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
Out of necessity ( I am developing video appliocation and need more than one webcam) - my cheap webcams did not come with driver for win 2000 - I switched to XP (SONY home edition).
I heve an external USB drive which holds all my development archive.
This drive and other USB devices worked and attached to my Windows 2000 via D-link hub and switch box ( I have two dev mashines) just fine and every time (on powerup).
Now I need to restart the XP often before the Disk Manager "registers" the external drive.
I tryed motherboard USB ports ( back and front)and PCI USB (5 ports) card. Same results.
I have these devices on the USB - 4 webcams, external drive,USB hub with serial port adapter( ham radio), scanner, MS wireless keboard / mouse.flash drive. All of them have some intermittent issues.
I guess the qestion is - how to debug USB and find out what is happenning here.
Once I got some message about "bandwith 70% usage" problem but I have no clue what to do with it.
It wants me to "unplug" device which is stupid because if it slows dow I do not care, my apps is bulid with checks for that.
Anyway, could someone kindly shed some light on this USB behavior.
Thanks for reading.
Vaclav
PS All of the devices worked on Win 2000, some of them are powered, it cannot be a power issue.
|
|
|
|
|
XP changed the way drives load in order to speed up the boot process. Now, you get a GUI up in about 4 seconds, thats completely useless for 26 seconds because none of the HW is attached yet, and explorer is locked waiting for the network to connect.
This is clearly an improvement over 2k, where you just waited 30 seconds for the GUI to come up...
Anyway, USB, where to start... OK, it sucks. Its the lamest, buggiest, lest observed standardwise, most indifferently supported at hub level protocol ever seen.
Advice? For gods sake dont use Vista, its worse. As for XP, what does device manager say when the USB drive has not been recognised?
"It is a remarkable fact that despite the worldwide expenditure of perhaps US$50 billion since 1990, and the efforts of tens of thousands of scientists worldwide, no human climate signal has yet been detected that is distinct from natural variation."
Bob Carter, Research Professor of Geology, James Cook University, Townsville
|
|
|
|
|
Thanks for reply.
I foolishly though it was only XP SP3 problem!
And it is not only on startup!
In addition of "ignoring" USB hub at statup I have also observed that unplugging / plugging directly connected ( not via D link) devices is also ignored by XP.
I needed to switch to XP so I can use Logitech webcams, no other reason.
The biggest problem is that I have wireless keyboard / mouse also connected via the D-link hub!
For all practical purposes my USB is useless in XP.
My personal opinion - MS is in hardware development business and should stay there.
I always say the reason that CD was "invented" was because it was impractical to load Windows using two dozen 3 1/2 floppies!
They obviously have no business in operating system development since they cannot get it together after all these years.
Too bad general public does not see that.
If Windows 2011 will make coffee they will all line up for it!
Vaclav
PS I'll check the Device Manager next time.
|
|
|
|
|
Yeah, check device manager. You will see a great big yellow exclamaitn mark on the fialed device.
Also, how does the HW work on other laptops? Becauae god knows USB host controllers can be very VERY variable in quality so dont blame MS, they just make the OS, its someone else made the HW...
"It is a remarkable fact that despite the worldwide expenditure of perhaps US$50 billion since 1990, and the efforts of tens of thousands of scientists worldwide, no human climate signal has yet been detected that is distinct from natural variation."
Bob Carter, Research Professor of Geology, James Cook University, Townsville
|
|
|
|
|
I've got a Dell laptop that has developed an odd problem. As I type, about halfway through a sentence I look up and realize that the cursor has jumped backwards several words, and the characters I'm typing are appearing in the middle of other words earlier in the line. It's completely random, and is not associated with any particular key. The backwards leap always stays within the same line of the display, as well; it doesn't VT to a previous line. It's as if the keyboard is generating a random CR withut the LF, except that it doesn't back up all the way to the left side of the line. It more often is a few characters in from the left side.
I have another identical laptop which is thoroughly dead, but before I go to the effort of trying to swap keyboards I thought I'd ask... Is this likely to be just a keyboard problem, or is it probable that this is something more insidious?
Will Rogers never met me.
|
|
|
|
|
Any chance you are (nearly) touching the touchpad/trackpad while typing? happens to me on occasion.
Staying in the same line is odd. Does it happen in multiple applications? If yes, write a little test app that logs all the keyboard [ADDED] and mouse [/ADDED] events, and match them up with your actual typing.
[ADDED] If need be, you can change the touchpad's sensitivity through the Mouse Control Panel. [/ADDED]
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles] Nil Volentibus Arduum
Please use <PRE> tags for code snippets, they preserve indentation, improve readability, and make me actually look at the code. modified on Thursday, December 16, 2010 7:28 PM
|
|
|
|
|
Luc Pattyn wrote: Any chance you are (nearly) touching the touchpad/trackpad while typing
None. After I noticed this happening, I took some time to experiment, typing very slowly and carefully with just one finger.
Luc Pattyn wrote: write a little test app that logs all the keyboard [ADDED] and mouse [/ADDED] events, and match them up with your actual typing.
That's a good thought... I'll give that a try next week. 
Will Rogers never met me.
|
|
|
|
|
I'm with Luc on this one. Every time I've experienced what you are describing, it's due to a stray finger touching the friggin' touch pad while typing.
CQ de W5ALT
Walt Fair, Jr., P. E.
Comport Computing
Specializing in Technical Engineering Software
|
|
|
|
|
See above. There's no chance of any accidental contact causing this. But the laptop was available - cheap - on eBay. Now I know why.
Will Rogers never met me.
|
|
|
|
|
OK, well, I recall that one other time I had some weird stuff going on, was when my ham radio interfered with the laptop. Isolating the cables took care of that. But in your case, I have no idea how to check and see if something like that is going on.
On the other hand, maybe you should just leave it as is. You could have a competition with DD on Fridays and if you ever do tyupe somtging spullid wrong, you have the perfect alibi.  
CQ de W5ALT
Walt Fair, Jr., P. E.
Comport Computing
Specializing in Technical Engineering Software
|
|
|
|
|
No, nothing weird going on around me at the time. Just a boring hotel room.
I wonder what DD would be willing to pay for an alibi? I have a laptop available that would certainly do the trick.
I need no alibi, as my drunken maunderings are always perfectly intelligible, if not particularly interesting. 
Will Rogers never met me.
|
|
|
|
|
Another idea: do you have anything else installed that can act as an input device? a bar code reader, a second keyboard, a gaming console, some kind of bluetooth device, ...? if so, disconnect all of them, and try again.
And finally: are you amidst strong electromagnetic fields, say in the middle of a power plant?
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
