Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Rest in Peace, Optane

specbranch.com

Post reply on HN
Print view View original

51–60 of 122 posts

Re: Rest in Peace, Optane

#51

Post by watersb »

tanelpoder wrote: post #24

Yeah I'm sad to see Optane go. I bought a few DCPMMs myself (and had to upgrade my workstation CPUs to support them) to test out the capabilities and maybe even write a little toy OS/KV-store engine that runs as a multithreaded process in Linux using mmap'ed Optane memory as its persistent datastore (never got to the latter part). The "do not use mmap for databases" argument would not apply here as you wouldn't cache…

> But Intel can probably still reuse lots of their cache coherency logic for external CXL.mem device access.

Holy cow, you nailed it: Intel can't get Optane fabrication cost down fast enough, and everyone is moving to CXL anyway, which presents its own latency challenges that tend to hide Optane performance advantage.

I wonder if you could leverage Optane's bit-level addressability in a shared memory pool scenario.

Sorry to see this tech go...

Re: Rest in Peace, Optane

#52

Post by tanelpoder »

eternityforest wrote: post #27

Why was everyone so obsessed with the idea of Optane as universal memory? From a consumer perspective, it was a fast small SSD that doesn't wear out. That was enough to justify it's value at the prices it went for(Or at least it seems like it, I never personally used it). It didn't need any radical rethink of computing to be worth it. I'm not even a fan of the nonvolatile RAM idea. Rebooting is often the first thing…

As it's part of the memory domain, not some external block device, the CPU caches can cache it (64B cache line sizes) and maintain cache coherency across CPUs, NUMA nodes (and soon with CXL 3.x ?) across memory pools shared by many servers. So your programming model will be just using CPU loads & stores to access the individual bytes you need directly from the Optane storage, instead of having to do some 512B or 4kB DMA or memcpy to RAM first and then access what you needed. And it's persistent, so you won't need to have a separate layer of code for persisting your "cache" of data... One can ditch the "database block I/O" and "buffer cache" notion and just treat Optane as byte-addressable RAM on a server that will never reboot or crash... Of course in reality your server can always catch fire or have a hardware malfunction, so you'd need to replicate your data to some distant enough backup or DR instance... which (at least in mission-critical database context) negates some of the value of persistent low-latency I/O... If you need to wait for 1+ms for the remote WAL write to be acknowledged from a different AZ or region anyway, then having an 1 us local Optane write is not gonna change the whole commit latency that radically.

Re: Rest in Peace, Optane

#53

Post by bravetraveler »

anonylizard wrote: post #39
mrb wrote: post #35

Earlier quoted context omitted.

" The theory that this market exists in any significant way never pans out in practice. " The transition from HDD to SSD technology is a counter example to your claim. It was a drop-in replacement (same SATA interface) and the tech significantly improved performance without any other software modification needed.

Didn't SSD's get a start with laptops and mobile devices? There the advantage of being shock resistant is massive and unsolvable with any other solution. For a superior but new tech to stick, it has to find a viable market to be first self-sustaining, only then can it attack larger and more lucrative markets (Back in those days, PC was probably more lucrative than mobile). Only when SSDs matured enough, did they star…

I wouldn't really say they started with mobile/laptops

They came in 2.5 inch form factor from the very beginning, the laptop shape... which fits pretty much anything. Desktops and servers use it.

The packaging was trivial. They were just cost prohibitive for anything more than an operating system drive.

This worked out to our benefit, the OS is where random access performance shines

It might be fair to say flash used in SSDs started in other applications... but it's important to remember that's also different flash.

It wasn't until the flash changed significantly that we got SSDs (durability), changing everything we once knew. Price, applications, etc.

Re: Rest in Peace, Optane

#54

Post by the8472 »

It's not just Optane. There have been other NVRAM technologies that were hyped as "fast as SRAM, persistent as HDDs, cheap and dense as DRAM" over the years such as FeRAM and MRAM. MRAM exists as a commercial product but it still fails to deliver on the cheap/dense part.

Re: Rest in Peace, Optane

#55

Post by watersb »

I have two 905p SSDs for ZFS cache devices. A two-way redundant point of failure, but these devices are way more durable than NAND flash storage.

A different form factor are the NVMe sticks. Given the relatively small capacity, the 118GB NVMe SSDs were not very expensive, and make ideal system drives for server applications.

But brute force seems to have defeated Optane: Intel's enterprise flash NAND SSDs are just super over-provisioned, retaining gobs of spare capacity that result in 8TB devices with one complete drive write per day durability, every day for five years guaranteed.

Re: Rest in Peace, Optane

#56

Post by FirmwareBurner »

znpy wrote: post #22
ritwikgupta wrote: post #13

Earlier quoted context omitted.

Apple may already be headed in that direction. They already have unified CPU and GPU RAM. It doesn’t seem far-fetched to imagine that they could unify persistent storage and memory.

Knowing apple’s marketing moves, they could definitely do that: just use a single number to describe memory. And then pretend it’s a big number.

I can already see it: base 128GB of total system unified memory for your files and data.

Re: Rest in Peace, Optane

#57

Post by lichtenberger »

What I'm missing the most is byte granular reading (in case of Optane it was 256 bytes, because of checksums, but that's fine granular enough).

It means it would be possible to read/write data in much more fine granular chunks (potentially saving a lot of storage space in some cases).

Re: Rest in Peace, Optane

#58

Post by gwd »

zinodaur wrote: post #14
hardware2win wrote: post #6

What was interesting about Optane was that it was kinda an attempt to get rid of disk and ram distinction. You had single device (optane stick) which served as both - ram and disk. After decades of current approach we finally could physically get rid of disks - hdds, ssds or even nvme sticks. https://www.theregister.com/2022/08/01/optane_intel_cancella... >Optane presented a radical, transformative technology but bec…

Yeah - the article was talking about mmap... but what i wanted was to not have to define persistence boundary. I wanted the entire in-memory state of my program to be persisted - perhaps even duplicated and moved elsewhere

I'll admit this sounded cool when I first heard about it; but it's actually a lot harder to program if you want to be able to recover from sudden power outages (which would be the main reason for having persistence in the first place).

Re: Rest in Peace, Optane

#59

Post by masklinn »

satiric wrote: post #48
ritwikgupta wrote: post #13

Earlier quoted context omitted.

Apple may already be headed in that direction. They already have unified CPU and GPU RAM. It doesn’t seem far-fetched to imagine that they could unify persistent storage and memory.

Well technically, Intel and AMD both use the regular system RAM for their integrated graphics VRAM, but I see what you mean.

Intel and AMD also do support unified memory for their integrated graphics. It’s been a while since you needed a statically cordoned-off area of main memory (“shared memory”) for the iGPU to work.

Consoles have been using unified memory since the 8th gen (PS4/XB1/Switch, kinda sorta even WiiU).

And NVidia has CUDA unified-memory slide decks going back at least 5 years.

Re: Rest in Peace, Optane

#60

Post by alpaca128 »

masklinn wrote: post #59
satiric wrote: post #48

Earlier quoted context omitted.

Well technically, Intel and AMD both use the regular system RAM for their integrated graphics VRAM, but I see what you mean.

Intel and AMD also do support unified memory for their integrated graphics. It’s been a while since you needed a statically cordoned-off area of main memory (“shared memory”) for the iGPU to work. Consoles have been using unified memory since the 8th gen (PS4/XB1/Switch, kinda sorta even WiiU). And NVidia has CUDA unified-memory slide decks going back at least 5 years.

The Xbox360 already had unified memory too. That gave it a slight edge compared to the PS3 in the long term because it was more flexible compared to a fixed 50:50 split.
Post reply on HN