_______               __                   _______
       |   |   |.---.-..----.|  |--..-----..----. |    |  |.-----..--.--.--..-----.
       |       ||  _  ||  __||    < |  -__||   _| |       ||  -__||  |  |  ||__ --|
       |___|___||___._||____||__|__||_____||__|   |__|____||_____||________||_____|
                                                             on Gopher (inofficial)
   URI Visit Hacker News on the Web
       
       
       COMMENT PAGE FOR:
   URI   What makes Intel Optane stand out (2023)
       
       
        ritcgab wrote 2 hours 21 min ago:
        All those nice numbers are just beaten by the unit cost. And the
        ecosystem is a mess.
       
        bluedino wrote 3 hours 3 min ago:
        Now do Intel's HBM/CPU Max
       
        myself248 wrote 3 hours 23 min ago:
        My kingdom for a MicroSD card with Optane inside. My dashcam wants it
        soooo badly.
       
        twotwotwo wrote 3 hours 43 min ago:
        One potential application I briefly had hope for was really good power
        loss protection in front of a conventional Flash SSD. You only need a
        little compared to the overall SSD capacity to be able to correctly
        report the write was persisted, and it's always running, so there's
        less of a 'will PLP work when we really need it?' question. (Maybe
        there's some use as a read cache too? Host RAM's probably better for
        that, though.) It's going to be rewritten lots of times, but it's
        supposed to be ready for that.
        
        It seems like there's a very small window, commercially, for new
        persistent memories. Flash throughput scales really cost-efficiently,
        and a lot is already built around dealing with the tens-of-microseconds
        latencies (or worse--networked block storage!). Read latencies you can
        cache your way out of, and writers can either accept commit latency or
        play it a little fast and loose (count a replicated write as safe
        enough or...just not be safe). You have to improve on Flash by enough
        to make it worth the leap while remaining cheaper than other approaches
        to the same problem, and you have to be confident enough in pulling it
        off to invest a ton up front. Not easy!
       
          wtallis wrote 3 hours 28 min ago:
          > One potential application I briefly had hope for was really good
          power loss protection in front of a conventional Flash SSD.
          
          That was never going to work out. Adding an entirely new kind of
          memory to your storage stack was never going to be easier or cheaper
          than adding a few large capacitors to the drive so it could save the
          contents of the DRAM that the SSD still needed whether or not there
          was Optane in the picture.
       
          hedora wrote 3 hours 31 min ago:
          Any decent SSD has capacitor (enterprise) or battery backed (phones)
          DRAM.  Therefore, a sync write is just “copy the data to an I/O
          buffer over PCIe”.
          
          For databases, where you do lots of small scattered writes, and lots
          of small overwrites to the tail of the log, modern SSDs coalesce
          writes in that buffer, greatly reducing write wear, and allowing the
          effective write bandwidth to exceed the media write bandwidth.
          
          These schemes are much less expensive than optane.
       
            olavgg wrote 1 hour 19 min ago:
            I have tried multiple enterprise SSD's, for sync writes. Nothing
            comes close to Optane Dimm, even Optane NVMe is 10x slower than
            PDIMMS.
            
   URI      [1]: https://forums.servethehome.com/index.php?threads/so-i-tes...
       
          zozbot234 wrote 3 hours 40 min ago:
          > It seems like there's a very small window, commercially, for new
          persistent memories. Flash throughput scales really cost-efficiently
          
          Flash is no bueno for write-heavy workloads, and the random-access
          R/W performance is meh compared to Optane.  MLC and SLC have better
          durability and performance, but still very mid.
       
        rkagerer wrote 3 hours 55 min ago:
        Did anyone ever see retention issues like this guy reported on one of
        his older models?
        
   URI  [1]: https://goughlui.com/2024/07/28/tech-flashback-intel-optane-3d...
       
          zozbot234 wrote 3 hours 50 min ago:
          That's data retention issues on the very first read-through of the
          media after sitting in cold storage for many years, with subsequent
          performance returning to normal.  It's definitely something to be
          aware of (and kudos to the blog poster for running that experiment)
          but worn-out NAND will behave a lot worse than that.
       
        exmadscientist wrote 4 hours 9 min ago:
        Around the time of Optane's discontinuation, the rumor mill was saying
        that the real reason it got the axe was that it couldn't be shrunk any,
        so its costs would never go down. Does anyone know if that's true? I
        never heard anything solid, but it made a lot of sense given what we
        know about Optane's fab process.
        
        And if no shrink was possible, is that because it was (a) possible but
        too hard; (b) known blocks to a die shrink; or (c) execs didn't want to
        pay to find out?
       
          georgeburdell wrote 3 hours 19 min ago:
          Flash has the same shrink problem.  And the solution for Optane was
          the same: go 3D
       
            exmadscientist wrote 3 hours 1 min ago:
            I don't think the shrink problem is at all the same for the two
            technologies. There are some really weird materials and production
            steps in Optane that are simply not present when making Flash
            cells.
       
          hedora wrote 3 hours 25 min ago:
          I think it was killed primarily because the DIMM version had a
          terrible programming API. There was no way to pin a cache line,
          update it and flush, so no existing database buffer pool algorithms
          were compatible with it.  Some academic work tried to address this,
          but I don’t know of any products.
          
          The SSD form factor wasn’t any faster at writes than NAND +
          capacitor-backed power loss protection.  The read path was faster,
          but only in time to first byte.  NAND had comparable / better
          throughput.  I forget where the cutoff was, but I think it was less
          than 4-16KB, which are typical database read sizes.
          
          So, the DIMMs were unprogrammable, and the SSDs had a “sometimes
          faster, but it depends” performance story.
       
            exmadscientist wrote 3 hours 11 min ago:
            The DIMMs were their own shitshow and I don't know how they even
            made it as far as they did.
            
            The SSDs were never going to be dominant at straight read or write
            workloads, but they were absolutely king of the hill at mixed
            workloads because, as you note, time to first byte was so low that
            they switched between read and write faster than anything short of
            DRAM. This was really, really useful for a lot of workloads, but
            benchmarkers rarely bothered to look at this corner... despite it
            being, say, the exact workload of an OS boot drive.
            
            For years there was nothing that could touch them in that corner
            (OS drive, swap drive, etc) and to this day it's unclear if the
            best modern drives still can or can't compete.
       
            myself248 wrote 3 hours 18 min ago:
            It sounds like they didn't do a good job of putting the DIMM
            version in the hands of folks who'd write the drivers just for fun.
            
            The read path is sort of a wash, but writes are still unequalled.
            NAND writes feel like you're mailing a letter to the floating
            gate...
       
              zozbot234 wrote 3 hours 15 min ago:
              Isn't this addressed by newer PCIe standards? Of course, even the
              "new" Optane media reviewed in OP is stuck on PCIe 4.0...
       
          zozbot234 wrote 4 hours 4 min ago:
          That's at least physically half-plausible, but it would be a terrible
          reason if true.  3.5 in. format hard drives can't be shrunk any, and
          their costs are correspondingly high, but they still sell - newer
          versions of NVMe even provide support for them.  Same for LTO tape
          cartridges.  Perhaps they expected other persistent-memory
          technologies to ultimately do better, but we haven't really seen
          this.
          
          Worth noting though that Optane is also power-hungry for writes
          compared to NAND.  Even when it was current, people noticed this. 
          It's a blocker for many otherwise-plausible use cases, especially re:
          modern large-scale AI where power is a key consideration.
       
            wtallis wrote 3 hours 33 min ago:
            >  3.5 in. format hard drives can't be shrunk any,
            
            You're looking at the entirely wrong kind of shrinking. Hard drives
            are still (gradually) improving storage density: the physical size
            of a byte on a platter does go down over time.
            
            Optane's memory cells had little or no room for shrinking, and
            Optane lacked 3D NAND's ability to add more layers with only a
            small cost increase.
       
        pgwalsh wrote 4 hours 17 min ago:
        Sure, they were expensive but they have great endurance and sustained
        read and write speeds. I use one in my car for camera recordings. I had
        gone through several other drives but this one has been going on 3 or 4
        years now without issue. I have a couple more in use too. It's a shame
        this tech is going away because it's excellent.
       
        rkagerer wrote 4 hours 22 min ago:
        My understanding is Optane is still unbeaten when it comes to latency.
        Has anyone examined its use as an OS volume, compared to today's
        leading SSD's?    I know the throughput won't be as high, but in my
        experience that's not as important to how responsive your machine feels
        as latency.
       
          dmayle wrote 1 hour 6 min ago:
          I run two 1.5TB Optanes in raid-0 with XFS (I picked them up for $300
          each on sale about two years ago).  These are limited to PCIE 3.0 x4
          (about 4GB/s max each).  I also have a 64GB optane drive I use as my
          boot drive.
          
          It's hard to tell you, because it's subjective, I don't swap back and
          forth between an SSD and the optane drives.  I have my old system,
          which has a 2TB Samsung 980 Pro NVME drive (PCIE 4.0 x4, or 8GB/s
          max) as root, and a Sabrent rocket 4 plus 4TB drive secondary (also
          PCIE 4.0), so I ran sysbench on both systems, so I could share the
          differences.  (Old system 5950X, new system 9950X3D).
          
          It feels snappier, especially when doing compilations...
          
          Sequential reads:
          I started with a 150GB fileset, but it was being served by the kernel
          cache on my newer system (256GB RAM vs 128GB on the old), so I
          switched to use 300GB of data, and the optanes gave me 5000 MiB/s for
          sequential read as opposed to 2800 MiB/s for the 980 Pro, and 4340
          MiB/s for the Rocket 4 Plus.
          
          Random writes alone (no read workload)
          The optane system gets 2184 MiB/s, the 980 Pro gets 32 MiB/s, and the
          Rocket 4 Plus gets 53 MiB/s.
          
          Mixed workload (random read/write)
          The optanes get 725/483 as opposed to 9/6 for the 980 Pro, and 42/28
          for the Rocket 4 Plus.
          
          2x1.5TB Optane Raid0:
              Prep time:
              `sysbench fileio --file-total-size=150G prepare`
              161061273600 bytes written in 50.41 seconds (3047.27 MiB/sec).
          
              Benchmark:
              `sysbench fileio --file-total-size=150G --file-test-mode=rndrw
          --max-time=60 --max-requests=0 run`
              WARNING: --max-time is deprecated, use --time instead
              sysbench 1.0.20 (using system LuaJIT 2.1.1741730670)
          
              Running the test with following options:
              Number of threads: 1
              Initializing random number generator from current time
          
              Extra file open flags: (none)
              128 files, 1.1719GiB each
              150GiB total file size
              Block size 16KiB
              Number of IO requests: 0
              Read/Write ratio for combined random IO test: 1.50
              Periodic FSYNC enabled, calling fsync() each 100 requests.
              Calling fsync() at the end of test, Enabled.
              Using synchronous I/O mode
              Doing random r/w test
              Initializing worker threads...
          
              Threads started!
          
              File operations:
              reads/s:              46421.95
              writes/s:              30947.96
              fsyncs/s:              99034.84
          
              Throughput:
              read, MiB/s:              725.34
              written, MiB/s:           483.56
          
              General statistics:
              total time:                 60.0005s
              total number of events:          10584397
          
              Latency (ms):
                   min:                     0.00
                   avg:                     0.01
                   max:                     1.32
                   95th percentile:                 0.03
                   sum:                 58687.09
          
              Threads fairness:
              events (avg/stddev):           10584397.0000/0.00
              execution time (avg/stddev):   58.6871/0.00
          
          2TB Nand Samsung 980 Pro:
              Prep time:
              `sysbench fileio --file-total-size=150G prepare`
              161061273600 bytes written in 87.15 seconds (1762.53 MiB/sec).
          
              Benchmark:
              `sysbench fileio --file-total-size=150G --file-test-mode=rndrw
          --max-time=60 --max-requests=0 run`
              WARNING: --max-time is deprecated, use --time instead
              sysbench 1.0.20 (using system LuaJIT 2.1.1741730670)
          
              Running the test with following options:
              Number of threads: 1
              Initializing random number generator from current time
          
              Extra file open flags: (none)
              128 files, 1.1719GiB each
              150GiB total file size
              Block size 16KiB
              Number of IO requests: 0
              Read/Write ratio for combined random IO test: 1.50
              Periodic FSYNC enabled, calling fsync() each 100 requests.
              Calling fsync() at the end of test, Enabled.
              Using synchronous I/O mode
              Doing random r/w test
              Initializing worker threads...
          
              Threads started!
          
              File operations:
              reads/s:              594.34
              writes/s:              396.23
              fsyncs/s:              1268.87
          
              Throughput:
              read, MiB/s:              9.29
              written, MiB/s:           6.19
          
              General statistics:
              total time:                 60.0662s
              total number of events:          135589
          
              Latency (ms):
                   min:                     0.00
                   avg:                     0.44
                   max:                    15.35
                   95th percentile:                 1.73
                   sum:                 59972.76
          
              Threads fairness:
              events (avg/stddev):           135589.0000/0.00
              execution time (avg/stddev):   59.9728/0.00
          
          4TB Sabrent Rocket 4 Plus:
              Prep time:
              `sysbench fileio --file-total-size=300G prepare`
              322122547200 bytes written in 152.39 seconds (2015.92 MiB/sec).
          
              Benchmark:
              `sysbench fileio --file-total-size=300G --file-test-mode=rndrw
          --max-time=60 --max-requests=0 run`
              WARNING: --max-time is deprecated, use --time instead
              sysbench 1.0.20 (using system LuaJIT 2.1.1741730670)
          
              Running the test with following options:
              Number of threads: 1
              Initializing random number generator from current time
          
              Extra file open flags: (none)
              128 files, 2.3438GiB each
              300GiB total file size
              Block size 16KiB
              Number of IO requests: 0
              Read/Write ratio for combined random IO test: 1.50
              Periodic FSYNC enabled, calling fsync() each 100 requests.
              Calling fsync() at the end of test, Enabled.
              Using synchronous I/O mode
              Doing random r/w test
              Initializing worker threads...
          
              Threads started!
          
              File operations:
              reads/s:              2690.28
              writes/s:              1793.52
              fsyncs/s:              5740.92
          
              Throughput:
              read, MiB/s:              42.04
              written, MiB/s:           28.02
          
              General statistics:
              total time:                 60.0155s
              total number of events:          613520
          
              Latency (ms):
                   min:                     0.00
                   avg:                     0.10
                   max:                     8.22
                   95th percentile:                 0.32
                   sum:                 59887.69
          
              Threads fairness:
              events (avg/stddev):           613520.0000/0.00
              execution time (avg/stddev):   59.8877/0.00
       
          aaronmdjones wrote 3 hours 35 min ago:
          I have a 16 GiB Optane NVMe M.2 drive in my router as a boot drive,
          running OpenWRT.
          
          It's so incredibly fast and responsive that the LuCI interface
          completely loads the moment I hit enter on the login form.
       
          hamdingers wrote 4 hours 0 min ago:
          > Has anyone examined its use as an OS volume, compared to today's
          leading SSD's?
          
          Late last year I switched from a 1.5tb Optane 905P to a 4tb WD Blue
          SN5000 NVMe drive in a gaming machine and saw improved load times,
          which makes sense given the read and write speeds are ~double. No
          observable difference otherwise.
          
          I'm sure that's not the use case you were looking for. I could
          probably tease out the difference in latency with benchmarks but
          that's not how I use the computer.
          
          The 905P is now in service as an SSD cache for a large media server
          and that came with a big performance boost but the baseline I'm
          comparing to is just spinning drives.
       
            exmadscientist wrote 3 hours 6 min ago:
            Unfortunately a gaming machine workload is so read-heavy that I
            wouldn't expect Optane to square up well.  Gaming is all about read
            speed and overall capacity. You need that heavy I/O mix, especially
            with low latency deadlines, to see gains from Optane. That limited
            target use case, coupled with ignorant benchmarking, always limited
            them.
       
            rkagerer wrote 3 hours 46 min ago:
            Thanks, that's helpful real-world feedback (not that I wouldn't
            also be interested in some synthetic benchmark comparisons from
            someone else).
       
              aggieNick02 wrote 2 hours 52 min ago:
              We benchmarked three of the popular Optane NVMe SSDs about three
              years ago. There was a short window when they were on clearance
              and a popular choice as a cache SSD in TrueNAS. [1] You can
              compare their benchmarks with the other almost 400 SSDs we've
              benchmarked. Most impressive is that three years later they are
              still the top random read QD1 performers, with no traditional
              flash SSD coming anywhere close: [2] They are amazing for how
              consistent and boring their performance is. Bit level access
              means no need for TRIM or garbage collection, performance doesn't
              degrade over time, latency is great, and random IO is not
              problematic.
              
   URI        [1]: https://pcpartpicker.com/forums/topic/425127-benchmarkin...
   URI        [2]: https://pcpartpicker.com/products/internal-hard-drive/be...
       
          speedgoose wrote 4 hours 16 min ago:
          I configured a hetzner ax101 bare metal server with a 480GB 3d xpoint
          ssd some years ago. It’s used as the boot volume and it seems fast
          despite the server being heavily over provisioned, but I can’t
          really compare because I don’t have a baseline without.
       
          rkagerer wrote 4 hours 17 min ago:
          Before people claim it doesn't matter due to OS write buffering, I
          should point out a) today's bloated software and the many-layered,
          abstracted I/O stack it's built on tends to issue lots of unnecessary
          flushes, b) read latency is just as important as write (if not
          moreso) to how responsive your OS feels, particularly if the whole
          thing doesn't fit in (or preload to) memory.
       
        gigatexal wrote 4 hours 39 min ago:
        I’m still sad they discontinued them. What’s the alternative now
        does anything come close?
       
          walterbell wrote 3 hours 13 min ago:
          Small sizes are on secondary market for ~$1/GB.
       
            zozbot234 wrote 3 hours 1 min ago:
            Which is a bargain compared to what DRAM costs today. If you just
            include the bare minimum of DRAM for a successful boot and
            immediately set up the entire "small" Optane drive as swap, that's
            a viable workstation-class system for comparative peanuts. You
            can't do this with NAND because the write workload of swap kills
            the media (I suppose it becomes viable if you monitor SMART wearout
            indicators and heavily overprovision the storage to leverage the
            drive's pSLC mode, but you're still treating $~0.10/GB hardware as
            a consumable and that will cost you) and of course you can't do it
            with spinning rust because the media is too slow.
       
        readitalready wrote 4 hours 40 min ago:
        These are absolute beasts for database servers, and definitely needs to
        make a comeback.
        
        They suck for large sequential file access, but incredible for small
        random access: databases.
       
        gozzoo wrote 4 hours 40 min ago:
        Maybe we can also mention the HP Memristor here.
       
          jamiek88 wrote 3 hours 19 min ago:
          Oh I was so excited for that. I devoured any news or blogs or rumours
          about that immediately!
       
        dangoodmanUT wrote 4 hours 44 min ago:
        Optane was crazy good tech, it way just too expensive at the time for
        mass adoption, but the benefits were so good.
        
        Looking at those charts, besides the DWPD it feels like normal NVMe has
        mostly caught up. I occassionally wonder where a gen 7/8(?) optane
        would be today if it caught on, it'd probably be nuts.
       
          exmadscientist wrote 4 hours 12 min ago:
          The actual strength of Optane was on mixed workloads. It's hard to
          write a flash cell (read-erase-write cycle, higher program voltage,
          settling time, et cetera). Optane didn't have any of that baggage.
          
          This showed up as amazing numbers on a 50%-read, 50%-write mix.
          Which, guess what, a lot of real workloads have, but benchmarks don't
          often cover well. This is why it's a great OS boot drive: there's so
          much cruddy logging going on (writes) at the same time as reads to
          actually load the OS. So Optane was king there.
       
            lvl155 wrote 2 hours 41 min ago:
            It’s the best OS drive especially p5800x.
       
          zozbot234 wrote 4 hours 38 min ago:
          > besides the DWPD it feels like normal NVMe has mostly caught up.
          
          So what you mean is that on the most important metric of them all for
          many workloads, Flash-based NVMe has not caught up at all.  When you
          run a write heavy workload on storage with a limited DWPD (including
          heavy swapping from RAM) higher performance actually hurts your
          durability.
       
        ece wrote 4 hours 50 min ago:
        Fabs are expensive and all, but maybe running a right-sized fab could
        have still been profitable at making optane for low-latency work that
        it was so good at. Even moreso with RAM prices as they are.
       
          downrightmike wrote 16 min ago:
          PE popped already, now Private credit is already popping, having been
          in too many bubbles, Datacenter bubble popped them. The whole system
          just isn't talking about it because the media isn't doing their job
          to report. There are no more good loans to make, the ROI is nil and
          the interest rate is spot plus 5.5%, and businesses don't want to pay
          10% interest on a loan. We're ina zombie cycle
       
        FpUser wrote 5 hours 9 min ago:
        I feel sorry about the situation. From my perspective Optane was a
        godsend for databases. I was contemplating building a system. Could've
        been a pinnacle of vertical scalability for cheap.
       
        amelius wrote 5 hours 11 min ago:
        For a good technical explanation at the physical level of a memory
        cell:
        
   URI  [1]: https://pcper.com/2017/06/how-3d-xpoint-phase-change-memory-wo...
       
        walterbell wrote 5 hours 12 min ago:
        Related: "High-bandwidth flash progress and future" (15 comments), [1]
        In an era of RAM shortages and quarterly price increases, Optane
        remains viable for swap and CPU/GPU cache.
        
   URI  [1]: https://news.ycombinator.com/item?id=46700384
       
          Weryj wrote 3 hours 4 min ago:
          I’ve been considering buying 8x64g models and setting them as equal
          priority swap disks (to mitigate the low throughput) for this exact
          reason.
       
            MrDrMcCoy wrote 1 hour 48 min ago:
            Can confirm doing so is awesome. Get some slightly bigger ones and
            partition them for additional use as zil. They're extremely
            satisfying to use, and depressing to remember that we'll never see
            their like again.
       
          newsclues wrote 4 hours 55 min ago:
          in an era of shortages, if there was an optane factory today ready to
          print money...
       
            walterbell wrote 4 hours 38 min ago:
            Secondary market surplus pricing (~$1/GB) value accrues to the
            buyer..
       
              zozbot234 wrote 4 hours 29 min ago:
              > (~$1/GB)
              
              Isn't that actually crazy good, even insane value for the
              performance and DWPD you get with Optane, especially with DRAM
              being ~$15/GB or so? I don't think ~$1/GB NAND is anywhere that
              good on durability, even if the raw performance is quite possibly
              higher.
       
          trollbridge wrote 4 hours 58 min ago:
          Yeah, I've wondered if we might see a revival of this kind of
          technology.
       
        ashvardanian wrote 5 hours 13 min ago:
        I don't have the inside scoop on Intel's current mess, but they
        definitely have a habit of killing off their coolest projects.
       
          brcmthrowaway wrote 1 hour 41 min ago:
          Realsense too
       
        hbogert wrote 5 hours 13 min ago:
        It stands out, because it didn't sell. Which is weird because there
        were some pretty big pros about using them. The latency for updating 1
        byte was crazy good. Some databases or journals for something like zfs
        really benefited from this.
       
          mort96 wrote 1 hour 50 min ago:
          I never understood what they're meant to do. Intel seemed to picture
          some future where RAM is persistent; but they were never close to
          fast enough to replace RAM, and the option to reboot in order to fix
          some weird state your system has gotten itself into is a feature of
          computers, not a problem to work around.
       
          thesz wrote 2 hours 31 min ago:
          In "databases and journals" you rarely update just one byte, you do a
          transaction that updates data, several indexes and metadata. All of
          that needs to be atomic.
          
          Power failure can happen in between any of "1 byte updates with crazy
          latencies." However small latency is, power failure is still faster.
          Usually, there is a write ahead or some other log that alleviates the
          problem, this log is usually written in streaming fashion.
          
          What is good, though, is that "blast radius" [1] of failure is
          smaller than usual - failed one byte write rarely corrupts more that
          one byte or cache line. SQLite has to deal with 512 (and even more)
          bytes long possible corruptions on most disks, with Optane it is not
          necessarily so. So, less data to copy, scan, etc.
          
   URI    [1]: https://sqlite.org/psow.html
       
          amluto wrote 4 hours 24 min ago:
          Intel did a spectacularly poor job with the ecosystem around the
          memory cells.  They made two plays, and both were flops.
          
          1. “Optane” in DIMM form factor. This targeted (I think) two
          markets.  First, use as slower but cheaper and higher density
          volatile RAM. There was actual demand — various caching workloads,
          for example, wanted hundreds of GB or even multiple TB in one server,
          and Optane was a route to get there. But the machines and DIMMs never
          really became available. Then there was the idea of using Optane
          DIMMs as persistent storage.  This was always tricky because the DDR
          interface wasn’t meant for this, and Intel also seems to have a lot
          of legacy tech in the way (their caching system and memory
          controller) and, for whatever reason, they seem to be barely capable
          of improving their own technology. They had multiple serious false
          starts in the space (a power-supply-early-warning scheme using NMI or
          MCE to idle the system, a horrible platform-specific register to poke
          to ask the memory controller to kindly flush itself, and the
          stillborn PCOMMIT instruction).
          
          2. Very nice NVMe devices. I think this was more of a failure of
          marketing.  If they had marketed a line of SSDs that, coupled with an
          appropriate filesystem, could give 99% fsync latency of 5
          microseconds and they had marketed this, I bet people would have
          paid. But they did nothing of the sort — instead they just threw
          around the term “Optane” inconsistently.
          
          These days one could build a PCM-backed CXL-connected memory mapped
          drive, and the performance might be awesome. Heck, I bet it
          wouldn’t be too hard to get a GPU to stream weights directly off
          such a device at NVLink-like speeds. Maybe Intel should try it.
       
            orion138 wrote 4 hours 13 min ago:
            One of the many problems was trying to limit the use of Optane to
            Intel devices. They should have manufactured and sold Optane memory
            and let other players build on top of it at a low level.
       
              amluto wrote 3 hours 50 min ago:
              > Optane memory
              
              Which “Optane memory”?  The NVMe product always worked on
              non-Intel. The NVDIMM products that I played with only ever
              worked on a very small set of rather specialized Intel platforms.
              I bet AMD could have supported them about as easily as Intel, and
              Intel barely ever managed to support them.
       
                wtallis wrote 3 hours 39 min ago:
                The consumer "Optane memory" products were a combination of
                NVMe and Intel's proprietary caching software, the latter of
                which was locked to Intel's platforms. They also did two
                generations of hybrid Optane+QLC drives that only worked on
                certain Intel platforms, because they ran a PCIe x2+x2 pair of
                links over a slot normally used for a single X2 or x4 link.
                
                Yes, the pure-Optane consumer "Optane memory" products were at
                a hardware level just small, fast NVMe drives that could be use
                anywhere, but they were never marketed that way.
       
                  myself248 wrote 3 hours 21 min ago:
                  Exactly. I happen to have all AMD sitting around here, and
                  buying my first Optane devices was a gamble, because I had no
                  idea if they'd work. Only reason I ever did, is they got
                  cheap at one point and I could afford the gamble.
                  
                  That uncertainty couldn't have done the market any favors.
       
                  amluto wrote 3 hours 28 min ago:
                  I feel like this is proving my point. You can’t read
                  “Optane” and have any real idea of what you’re buying.
                  
                  Also… were those weird hybrid SSDs even implemented by
                  actual hardware, or were they part of the giant series of
                  massive kludges in the “Rapid Storage” family where some
                  secret sauce in the PCIe host lied to the OS about what was
                  actually connected so an Intel driver could replace the
                  OS’s native storage driver (NVMe, AHCI, or perhaps
                  something worse depending on generation) to implement all the
                  actual logic in software?
                  
                  It didn’t help Intel that some major storage companies
                  started selling very, very nice flash SSDs in the mean time.
       
                    wtallis wrote 3 hours 20 min ago:
                    > were those weird hybrid SSDs even implemented by actual
                    hardware, or were they part of the giant series of massive
                    kludges
                    
                    They were definitely part of the series of massive kludges.
                    But aside from the Intel platforms they were marketed for,
                    I never found a PCIe host that could see both of the NVMe
                    devices on the drive. Some hosts would bring up the x2 link
                    to the Optane half of the drive, some hosts would bring up
                    the x2 link to the QLC half of the drive, but I couldn't
                    find any way to get both links active even when the drive
                    was connected downstream of a PCIe switch that definitely
                    had hardware support for bifurcation down to x2 links. I
                    suspect that with appropriate firmware hacking on the host
                    side, it may have been possible to get those drives fully
                    operational on a non-Intel host.
       
          cogman10 wrote 4 hours 27 min ago:
          IMO, the reason they didn't sell is the ideal usage for them is
          pairing them with some slow spinning disks.  The issue Optane had is
          that SSD capacity grew dramatically while the price plummeted.    The
          difference between Optane and SSDs was too small.  Especially since
          the M.2 standard  proliferated and SSDs took advantage of PCI-E
          performance.
          
          I believe Optane retained a performance advantage (and I think even
          today it's still faster than the best SSDs) but SSDs remain good
          enough and fast enough while being a lot cheaper.
          
          The ideal usage of optane was as a ZIL in ZFS.
       
            exmadscientist wrote 4 hours 14 min ago:
            > The ideal usage of optane was as a ZIL in ZFS.
            
            It was also the best boot drive money could buy. Still is, I think,
            though other comments in the thread ask how it compares against
            today's best, which I'd also love to see.
       
              gozzoo wrote 3 hours 49 min ago:
              This concept was very popular back in the days when computers
              used to boot from HDD, but now it doesn't make much sense. I
              wouldn't notice If my laptop boots for 5 sec instead of 10.
       
                exmadscientist wrote 3 hours 19 min ago:
                At the time of their introduction Optane drives were noticeably
                faster to boot your machine than even the fastest available
                Flash SSD. So in a workstation with multiple hard drives
                installed anyway, buying one to boot off of made decent sense.
                
                If they had been cheaper, I think they'd have been really,
                really popular.
       
            bushbaba wrote 4 hours 21 min ago:
            Not just capacity but SSD speeds also improved to the point it was
            good enough for many high memory workloads.
       
            zozbot234 wrote 4 hours 22 min ago:
            That may have been the ideal usage back in the day, but ideal usage
            now is just for setting up swap.  Write-heavy workloads are king
            with Optane, and threshing to swap is the prototypical example of
            something that's so write-heavy it's a terrible fit for NAND. 
            Optane might not have been "as fast as DRAM" but it was plenty
            close enough to be fit for purpose.
       
              mort96 wrote 1 hour 47 min ago:
              That would be fine if I could put it in an M.2 slot. But all my
              computers already have RAM in their RAM slots, and even if I had
              a spare RAM slot, I don't know that I'd trust the software stack
              to treat one RAM slot as a drive...
              
              And their whole deal was making RAM persistent anyway, which
              isn't exactly what I want.
       
                zozbot234 wrote 1 hour 43 min ago:
                Optane M.2-format hardware exists.
       
                  mort96 wrote 1 hour 20 min ago:
                  Interesting, all I ever saw advertised was that weird
                  persistent kinda slow RAM stick. Does the M.2 version just
                  show up as a normal block device or is that too trying to be
                  persistent RAM?
       
          ksec wrote 4 hours 51 min ago:
          >Which is weird....
          
          It isn't weird at all. I would be surprised if it ever succeed in the
          first place.
          
          Cost was way too high. Intel not sharing the tech with others other
          than Micron. Micron wasn't committed to it either, and since unused
          capacity at the Fab was paid by Intel regardless they dont care. No
          long term solution or strategy to bring cost down. Neither Intel or
          Micron have a vision on this. No one wanted another Intel only tech
          lock in. And despite the high price, it barely made any profits per
          unit compared to NAND and DRAM which was at the time making historic
          high profits. Once the NAND and DRAM cycle went down again cost /
          performance on Optane wasn't as attractive. Samsung even made some
          form of SLC NAND that performs similar to Optane but cheaper, and
          even they end up stopped developing for it due to lack of interest.
       
            amluto wrote 1 hour 23 min ago:
            A ways back, I wrote a sort of database that was memory-mapped-file
            backed (a mistake, but I didn’t know that at the time), and I
            would have paid top dollar for even a few GB of NVDIMMs that could
            be put in an ordinary server and could be somewhat
            straightforwardly mounted as a DAX filesystem. I even tried to do
            some of the kernel work.  But the hardware and firmware was such a
            mess that it was basically a lost cause.  And none of the tech ever
            seemed to turn into an actual purchasable product. I’m a bit
            suspicious that Intel never found product-market fit in part
            because they never had a credible product on the NVDIMM side.
            
            Somewhere I still have some actual battery-backed DIMMs (DRAM plus
            FPGA interposer plus awkward little supercapacitor bundle) in a
            drawer. They were not made by Intel, but Intel was clearly using
            them as a stepping stone toward the broader NVDIMM ecosystem. They
            worked on exactly one SuperMicro board, kind of, and not at all if
            you booted using UEFI.    Rebooting without doing the magic handshake
            over SMBUS [0] first took something like 15 minutes, which was not
            good for those nines of availability.
            
            [0] You can find my SMBUS host driver for exactly this purpose on
            the LKML archives. It was never merged, in part, because no one
            could ever get all the teams involved in the Xeon memory controller
            to reach any sort of agreement as to who owned the bus or how the
            OS was supposed to communicate without, say, defeating platform
            thermal management or causing the refresh interval to get out of
            sync with the DIMM temperature, thus causing corruption.
            
            I’m suspicious that everything involved in Optane development was
            like this.
       
            deepsquirrelnet wrote 4 hours 2 min ago:
            I worked at Micron in the SSD division when Optane (originally
            called crosspoint “Xpoint”) was being made. In my mind, there
            was never a real serious push to productize it. But it’s not
            clear to me whether that was due to unattractive terms of the joint
            venture or lack of clear product fit.
            
            There was certainly a time when it seemed they were shopping for
            engineers opinions of what to do with it, but I think they quickly
            determined it would be a much smaller market anyway from ssds and
            didn’t end up pushing on it too hard. I could be wrong though,
            it’s a big company and my corner was manufacturing and not
            product development.
       
              rjsw wrote 1 hour 31 min ago:
              A friend was working at Micron on a rackmount network server with
              a lot of flash memory, I didn't ask at the time what kind of
              flash it used. The project was cancelled when nearly finished.
       
              chrneu wrote 3 hours 31 min ago:
              I worked at Intel for a while and might be able to explain this.
              
              There were/are often projects that come down from management that
              nobody thinks are worth pursuing. When i say nobody, it might not
              just be engineers but even say 1 or 2 people in management who
              just do a shit roll out. There are a lot of layers of Intel and
              if even one layer in the Intel Sandwich drag their feet it can
              kill an entire project. I saw it happen a few times in my time
              there. That one specific node that intel dropped the ball on kind
              of came back to 2-3 people in one specific department, as an
              example.
              
              Optane was a minute before I got there, but having been excited
              about it at the time and somewhat following it, that's the vibe I
              get from Optane. It had a lot of potential but someone screwed it
              up and it killed the momentum.
       
                empiricus wrote 2 hours 58 min ago:
                This is actually insane. Do you mean 2-4 people in one
                department basically killed Intel? Roll to disbelief.
       
                  LASR wrote 1 hour 33 min ago:
                  Yes this is pretty common in large enterprise-ey tech
                  companies that are successful. There are usually a small
                  group of vocal members that have a strong conviction and
                  drive to make a vision a reality. This is contrary to popular
                  belief that large companies design by committee.
                  
                  Of course it works exceptionally well when the instinct turns
                  out to    be right. But can end companies if it isn’t.
       
                  wtallis wrote 2 hours 44 min ago:
                  It's somewhat plausible that a small group of people in one
                  department were responsible for the bad bets that made their
                  10nm process a failure. But it was very much a group effort
                  for Intel to escalate that problem into the prolonged
                  disaster. Management should have stopped believing the
                  undeliverable promises coming out of their fab side after a
                  year or two, and should have started much sooner to design
                  chips targeting fab processes that actually worked.
       
                osnium123 wrote 3 hours 18 min ago:
                Are you referring to the Intel 10nm struggles in your reference
                to 2-3 people?
       
            jauntywundrkind wrote 4 hours 31 min ago:
            Cost was fantastically cheap, if you take into account that Optane
            is going to live >>10x longer than a SSD.
            
            For a lot of bulk storage, yes, you don't have frequently changing
            data. But for databases or caches, that are under heavy load,
            optane was not only far faster, but if looking at life-cycle costs,
            way way less.
       
              mapt wrote 1 hour 9 min ago:
              Write endurance of the drive would be measured in TBW, and TLC
              flash kept adding enough 3D layers to stay cheap enough, quickly
              enough, that Optane never really beat their pricing per TBW  to
              make a practical product.
              
              I have to wonder if it isn't usable for some kind of specialized
              AI workflow that would benefit from extremely low latency reads
              but which is isn't written often, at this point.  Perhaps
              integrated in a GPU board.
       
                zozbot234 wrote 32 min ago:
                Optane practical TBW endurance is way higher than that of even
                TLC flash, never mind QLC or PLC which is the current standard
                for consumer NAND hardware.  It even seems to go way beyond
                what's stated on the spec sheet. However, while Optane excels
                for write-heavy workloads (not read-heavy, where NAND actually
                performs very well) these are also power-hungry which is a
                limitation for modern AI workflow.
       
                jauntywundrkind wrote 33 min ago:
                The extra capacity of modern SSD is a good point, especially
                now that we have 100TB+ SSD.
                
                But Optane still offered 100 DWPD (drive writes per day), up to
                3.2TB. Thats still just so many more DWPD than flash ssd. A
                Kioxia    CM8V for example will do 12TB at 3 DWPD. The net TBW is
                still 10x apart.
                
                You can get back to high endurance with SLC drives like the
                Solidigm p7-p5810, but you're back down to 1.6TB and 50 DWPD,
                so, 1/4 the Intel P5800X endurance, and worse latencies. I
                highly suspect the drive model here is a homage, and in spite
                of being much newer and very expensive, the original is still
                so much better in so many ways. [1] You also end up paying for
                what I assume is a circa six figure drive, if you are
                substituting DWPD with more capacity than you need. There's
                something elegant about being able to keep using your cells,
                versus overbuying on cells with the intent to be able to rip
                through them relatively quickly.
                
   URI          [1]: https://www.solidigm.com/content/solidigm/us/en/produc...
       
              wtallis wrote 3 hours 44 min ago:
              Optane was in the market during a time when the mainstream trend
              in the SSD industry was all about sacrificing endurance to get
              higher capacity. It's been several years, and I'm not seeing a
              lot of regrets from folks who moved to TLC and QLC NAND, and
              those products are more popular than ever.
              
              The niche that could actually make use of Optane's endurance was
              small and shrinking, and Intel had no roadmap to significantly
              improve Optane's $/GB which was unquestionably the technology's
              biggest weakness.
       
          zozbot234 wrote 4 hours 55 min ago:
          Optane didn't sell because they focused on their weird persistent
          DIMM sticks, which are a nightmare for enterprise where for many
          ordinary purposes you want ephemeral data that disappears as soon as
          you cut power. Thet should have focused on making ordinary storage
          and solving the interconnect bandwidth and latency problems
          differently, such as with more up-to-date PCIe standards.
       
            hrmtst93837 wrote 2 hours 7 min ago:
            PCIe was a bottleneck in consumer boxes, but that wasn't the whole
            problem. Optane's low latency and write endurance looked great on
            paper, yet once you put it behind SSD controllers and file systems
            built around NAND assumptions, a lot of the upside got shaved off
            before users ever saw it.
            
            "Just make it a faster SSD" was never a business. The DIMMs were
            weird, sure, but the bigger issue was that Optane made the most
            sense when software treated storage and memory as one tier, and
            almost nobody was going to rewrite kernels, DBs, and apps for a
            product that cost more than flash and solved pain most buyers
            barely felt.
       
            jauntywundrkind wrote 4 hours 35 min ago:
            I don't think that would be my main complaint. Sticking optane in a
            dimm was just awkward as hell. You now have different bits of
            memory with very different characteristics, & you lose a ton of
            bandwidth.
            
            If CXL was around at the time it would have been such a nice fit,
            allowing for much lower latency access.
            
            It also seems like in spite of the bad fit, there were enough
            regular options drives, and they were indeed pretty incredible.
            Good endurance, reasonable price (and cheap as dirt if you consider
            that endurance/lifecycle cost!), some just fantastic performance
            figures. My conclusion is that alas there just aren't many people
            in the world who are serious about storage performance.
       
              tayo42 wrote 3 hours 54 min ago:
              Can Linux differentiate that different dimms are different? Or
              does it see it all as one big memory space still?
       
                wmf wrote 1 hour 27 min ago:
                Yes, Linux was aware of the difference via ACPI tables.
       
          p-e-w wrote 5 hours 3 min ago:
          Optane was a victim of its own hype, such as “entirely new
          physics”, or “as fast as RAM, but persistent”. The reality felt
          like a failure afterwards even though it was still revolutionary,
          objectively speaking.
       
          epistasis wrote 5 hours 3 min ago:
          When most people are running databases on AWS RDS, or on ridiculous
          EBS drives with insanely low throughput and latency, it makes sense
          to me.
          
          There are very few applications that benefit from such low latency,
          and if one has to go off the standard path of easy, but slow and
          expensive and automatically backup up, people will pick the ease.
          
          Having the best technology performance is not enough to have product
          market fit. The execution required from the side of executives at
          Intel is far far beyond their capability. They developed a platform
          and wanted others to do the work of building all the applications.
          Without that starting killer app, there's not enough adoption to
          build an ecosystem.
       
            amluto wrote 3 hours 25 min ago:
            > There are very few applications that benefit from such low
            latency
            
            Basically any RDBMS?  MySQL and Postgres both benefit from high
            performance storage, but too many customers have moved into the
            cloud where you can’t get NVMe-like performance for durable
            storage for anything remotely close to a worthwhile price.
       
              epistasis wrote 3 hours 7 min ago:
              I'm saying that there are very few downstream applications that
              use databases that benefit from reducing latency beyond the slow
              performance of the cloud. Running your database on VMs or
              baremetal gives better performance, but almost no applications
              built on databases bother to do it.
       
          bombcar wrote 5 hours 4 min ago:
          It feels like everyone figured out what to do with them and how just
          about when they stopped making them.
       
            timschmidt wrote 4 hours 56 min ago:
            Same for the Larabee / Knights architecture.  Would sure be fun to
            play around with a 500 core Knights CPU with a couple TB of optane
            for LLM inference.
            
            Intel's got an amazing record of axing projects as soon as they've
            done the hard work of building an ecosystem.
       
              zozbot234 wrote 4 hours 51 min ago:
              > 500 core
              
              The newest fully E-core based Xeon CPUs have reached that figure
              by now, at least in dual-socket configs.
       
                timschmidt wrote 4 hours 45 min ago:
                Yup.  And high end GPU compute now has on-package HBM like
                Knight's had a decade ago, and those new Intel CPUs are finally
                shipping with AVX reliably again.  We lost a decade for
                workloads that would benefit from both.
       
       
   DIR <- back to front page