Romex Software Forum

If you are loosing counters randomly there is a fix I found and presented here:
viewtopic.php?f=35&t=2582&p=7341#p7341
Never have had problems since I applied this.

But when you encrypt a partition and then cache it with PrimoCache (L2), is the data encrypted on the cache volume too? I would have run a test on that too even if it wasn't a part of my initial concern, but Primocache 1.01 was unable to recognise TrueCrypt volumes on my system. Use encrypted file ...

But when you encrypt a partition and then cache it with PrimoCache (L2), is the data encrypted on the cache volume too?

Since SATA channels are fully capable of independently and asynchronously reading and writing to different channels, I see no reason for any L2 delay at all.(...). Well, it is not so simple. If one wants to read from the volume and write to the cache at the same time the speed of such operations sh...

There is a registry value (not present on default) which prevents PerfLib from disabling counters (for instance if internal checking fails). To prevent auto disabling counters one has to do the following: Locate in the registry: HKLM\Software\Microsoft\Windows NT\CurrentVersion\Perflib Add the follo...

If you have 2GB of defer waiting it means you have sequential transfer or Latency set to very high.
Detecting if the transfer is sequential or random is very hard and does not work correctly in case of alternative cache solutions (tested by myself).
Where SSD's perform best RAM is even better.

But what's the point of L2 cache for write operations if you can use L1 for the same thing and with better performance.
You want to accelerate big sequential writing? You'll wear out ssd quickly that way.

Redundant Arrays (RAID) include verification subroutines only as optional maintenance procedure. It is not build as a part of the storage system per se. Too much performance penalty. Besides if inconsistency occurs how one is able to tell, which source is corrupted (cache or drive)?

That would increase the Hit Ratio considerably. But I'm willing to propose configurable storage for L1 saving. Not all of us use L1 in conjunction with L2. The saving system should cover all scenarios for instance separate L1 and L1/L2 caches for different partitions etc.

That's because windows uses its own disk cache and the second run usually comes from it. It seems that Primo do not interfere in this cache. It intercepts hard drive calls. No calls -> no led -> no count. I say, don't worry, as windows cache is in RAM so the fastest source available. But it happens ...