Several ways to get allocation unit or cluster size:
1. Right-click on the drive in My Computer or an Explorer window, select Format from the context menu. The allocation unit size will be listed on the resulting dialog.
2. Most (all?) disk defragmenting programs will tell you the allocation unit size.
3. If you run chkdsk (from the command prompt or at boot time) on a drive the third from last line in the results screen tells you the allocation unit size.
If you originally formatted that drive as NTFS and did not choose something other than default allocation unit size, then its allocation unit size will be 4,096 bytes. If you used the CONVERT utility to convert the drive from FAT or FAT32, then the cluster size will be 512 bytes, which would cause a lot of extra drive activity at startup, and perhaps at shutdown. But I wouldn't expect it to cause the behavior you're seeing continuously without ever stopping. (That is what's happening, isn't it?)
I suppose this could be a characteristic behavior of that specific drive and controller combination. I'm not familiar with the Cheetah, so I just don't know about that.
I don't think you told me yet whether or not you've altered the default pagefile settings. That could possibly have some bearing on this matter.
I think you'll have to sort through the processes you have running. My guess is that some application, or at least some specific setting on an application, is causing this behavior. The process list can be misleading because it's possible for a process to use minimal CPU time while still being somewhat disk-intensive (and vice versa). There are ways of monitoring disk use by process in NT and W2K, but they are somewhat involved to set up and analyze and are also prone to sap performance, so I'd start by trying a process of elimination on some of those processes you have running.
First things first, I'd go back over the previous list of items, particularly the anti-virus and communications software, to make sure that some setting or scheduled behavior isn't causing the symptoms. I also see several processes in your list which are either unfamiliar to me or are not running on any of the W2K Pro machines sitting in front of me. They include icq, wcmdmgr, devldr32, alertsvc, symtray, dhmgr, pdesk, poproxy, navapsvc, npssvc, navapw32. Yes, I know that ICQ and Norton utilities of various kinds are widely used. One look at ICQ and AIM and their use of memory has been enough to convince me I don't need them. And Norton ANYTHING is notorious for excessive (IMHO) resource use. Remember that anything that is memory-intensive can cause frequent swapping to / from the pagefile, particularly under certain circumstances. I notice in that list that pdesk has used a fair amount of processor time. What is it? I suggest that you look at the applications / utilities that are responsible for each of these processes. Any continuous monitoring process like a "disk doctor" or an aftermarket application or system monitor could cause behaviors like the one you are seeing. You could try removing or disabling all non-standard processes to see if the behavior goes away. If it does, re-enable the processes one-at-a-time to see which one is the culprit.
You could also communicate with other users of the Cheetah to see if they see the behavior. If this is just the way the Cheetah behaves, there's no sense in wasting time going through all the other motions. However, if this has a serious impact on performance I would be surprised if it were normal behavior. I don't know the driver personally, but I do know that most people who have them seem to be happy with them.
Regards,
Jim
Facebook
Twitter