Сhange mode - current cache: failed!

[sokolov]

I am sorry my language!

Note and problem..:

1. PrimoCache 1.0.1 - stable = it's problem no in Windows XP 64 bit (!)
2. I am create screenshot -> paste in editor..
3. I am edit working current cache from primocache -> edit not responding! Bugs!
4. I create new screenshot and current problem and send this forum.. )
5. I close process editor.

I am sorry crazy language, no found problem in current edit mode working cache.

99 percent it's no in Windows XP 64 All program in run: 1. VMWare Player (2 guest), Outpost Firewall, Aimp, Mozilla, Messenger.

If possible - answer the first question and the second problem.
Thank you!

PrimoCache 1.0.1 vs 2.0?: 2.0 is speed, yes, i think you need to work on the algorithm idle-flush/intelligetn (other i not testing).

[Support]

I am edit working current cache from primocache

If I understand correctly, do you mean the changing the cache configuration? As your post title said, I think you changed write mode. This will trigger the flushing process. In your case, it would write about 2GB data into the disk. This busy disk write opeation may cause some applications cannot respond in time.
I have a question about this problem. Did editor resume to respond after Defer-Write flushing was finished?

For your second question, Free Cache has nothing to do with Cache Hit Rate. See
http://www.romexsoftware.com/en-us/prim ... mance.html

[sokolov]

I have a question about this problem. Did editor resume to respond after Defer-Write flushing was finished?

NO

I am no foolish

[Support]

Thanks!

So there is a 1% chance the problem happens, right?

[sokolov]

almost surely.... ~2-3%

PS: I am current testin 2.0

[InquiringMind]

I've encountered a slightly different issue - changing mode is allowed but does not take effect until system restart, e.g. if I set up a read-only cache, then change it to read/write, I receive the "updated" dialogue message and everything indicates read/write but it performs as if still in read-only mode (checked using CDM benchmark) with no write caching until the system is restarted.

Not a major issue and just rewording the dialog to note that a restart may be needed could suffice (using WinXP SP3 with 18GB RAM, L1 cache using ~7GB invisible memory, no L2 with Primo Ramdisk using 8GB more invisible memory).

[Support]

@InquiringMind
Checked again, we haven't seen such issue. Can you give me more details on how you check with CDM?

[InquiringMind]

Done some more testing and it seems the error was on my end - apologies. I experimented with read-only versus read/write (for the write amplification issue discussed elsewhere) and I forgot to enable defer writes - that was the cause of the results I received.

[Support]