Romex Software Forum

You say: "Each cache block stores contiguous address sector data of corresponding size." Can you clarify what you mean by "corresponding size" with a simple example? By default for large disks, Windows will format NTFS with 4k block size. So if I understand you correctly, when Wi...

Regarding Ultracopier, yes, it would only limit the rate of the input to the cache - it knows nothing of the cache and vice versa. Still, that should help indirectly if you set a low enough latency in PrimoCache, especially with a very low rate set in Ultracopier, because it would only write in shor...

Thanks for clarifying. Re. the 256K block size, would you kindly briefly explain where it comes into play and the advantages of a bigger or smaller block size. I saw in the documentation that a smaller block size would cause PrimoCache to use more memory, so setting that aside, what is the effect ot...

Interesting request. I can see two circumstances where this would be useful: 1. When you want to read from the disks while they're being written to. 2. For older systems with slower PCIe bus, extended writes to disks may slow down the whole system considerably while they are ongoing. While you're wa...

I have a few questions about cache flushing: With both L1 & L2 configured (for the same underlying disk) with Defer-Write, how is the flushing of the L1 and L2 cache coordinated at the latency interval to avoid a simultaneous write for both? Also, in the case where, say, the L1 cache flush happe...

Can you just clarify "combining write requests for continuous address"? Thanks.

Got it about the reads. Thanks. If you don't mind, would you kindly clarify further on the writes? It might help me better understand the occasional performance problems I'm observing and figure out what to do about them. So: with 64kb block size in underlying hard drive, 512kb PrimoCache block size...

It's the "vice versa" I was wondering about: without read activity, how does PrimoCache determine what to populate with? For example, if PrimoCache gets only one 4k read request, after fetching that request will PrimoCache populate the whole 512kb block containing that 4k at the gather int...

Interesting. If read requests and L2 population are mostly independent, how does PrimoCache determine what it should populate the read cache with? If there is no read activity, will it still populate? Also, does the fetching only the data Windows/App requests (as opposed to the whole block) also app...

To clarify my question: when the app does a read of data that is not cached, say for example 4kb at a time, would PrimoCache retrieve the 512kb and then if the next 4kb read from the app addresses data in the same PrimoCache block that was just retrieved it would not need to go to disk? Also, how wo...