Jump to content
7 minutes ago, Pizzaguy94 said:

reads infinite?

Yes.

mY sYsTeM iS Not pErfoRmInG aS gOOd As I sAW oN yOuTuBe. WhA t IS a GoOd FaN CuRVe??!!? wHat aRe tEh GoOd OvERclok SeTTinGS FoR My CaRd??  HoW CaN I foRcE my GpU to uSe 1o0%? BuT WiLL i HaVE Bo0tllEnEcKs? RyZEN dOeS NoT peRfORm BetTer wItH HiGhER sPEED RaM!!dId i WiN teH SiLiCON LotTerrYyOu ShoUlD dEsHrOuD uR GPUmy SYstEm iS UNDerPerforMiNg iN WarzONEcan mY Pc Run WiNdOwS 11 ?woUld BaKInG MY GRaPHics card fIX it? MultimETeR TeSTiNG!! aMd'S GpU DrIvErS aRe as goOD aS NviDia's YOU SHoUlD oVERCloCk yOUR ramS To 5000C18! jellYfIn Client siDE TRanscoDinG

Link to comment
https://linustechtips.com/topic/1353989-total-host-reads/#findComment-14853600
Share on other sites

Link to post
Share on other sites

Host Write Commands is NOT that 600 TBW value. 

 

It's exactly what it says, how many write commands were received by the SSD.  The SSD writes some stuff into memory which is in SLC mode, then later moves that stuff to proper TLC memory. Sometimes if the data is erased soon after it's written, the data doesn't even get to be moved into a more performant location in the TLC memory. 

The memory area in SLC mode has much higher endurance, while the TLC area has less. 

 

Also, it's not the writing that causes "wear and tear" on the SSD, it's the erasing. Memory is organized in chunks of let's say 24-32 MB, and in these chunks there's "pages" of let's say 4096 bytes.  The SSD can write such 4096 byte pages but can't overwrite or erase the pages once they're written. In order to write new stuff, the whole chunk of 24-32 MB or whatever has to be erased and that erase process causes damage to that chunk of memory. QLC supports less than 1000 erases,  TLC supports around 1000-3000 erases, MLC could do up to around 6000-8000 erases

 

So unlike mechanical drives which can always overwrite data, if you want to overwrite something (let's say you edit a sentence in a TXT document) the SSD simply marks that specific 4096 byte page as "can be erased at some point in the future" and copies the content with the edits into a new page in some other chunk of memory that's writable.  When enough such pages marked as "can be erased" are found in a chunk, the SSD moves the remaining pages with still valid content from that chunk in other chunks and then erases the whole chunk to make it writable again and increments the "erases" counter for that chunk. 

 

So, the total amount of data written to NAND (flash memory) can actually be bigger than the Host writes (because the SSD silently copies pages around to make erasing such chunks possible when needed)

 

The 600 TBW is an estimation of how much you can write until there's too many flash memory cells that can no longer be reliably erased to accept new data and store that data reliably. Once such memory areas reach an erase counter high enough, they're marked as read only and data is no longer written to those areas. 

 

Link to comment
https://linustechtips.com/topic/1353989-total-host-reads/#findComment-14853613
Share on other sites

Link to post
Share on other sites

Reads are effectively infinite, although not precisely without impact. Reading the same stale data over and over again can cause read disturb and eventually the controller must refresh (rewrite) this flash, so you get more writes. However, once rewritten it's like it never happened. So it's not something to be worried about. For the record, typical consumer usage is 70% reads and 30% writes.

Link to comment
https://linustechtips.com/topic/1353989-total-host-reads/#findComment-14855737
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×