Live data from Hacker News

What makes Intel Optane stand out (2023)

blog.zuthof.nl

151–160 of 164 posts

Re: What makes Intel Optane stand out (2023)

#151
post #137

Did it ever have real input costs which justified its pricing or was it totally a synthetic price to cover an imagined value above base, and some IPR? I may be wrong, but I tend to think the fab cost has very little to do with the complexity, but not nothing, and the outcome cost is very unrelated to price at a viable yield. If optane failed, it was probably because Intel didn't like the economics of pricing it to su…

I still think it has a lot of potential, and is the logical next storage system, because of its performance advantages over NAND flash. Because input costs can come down.

"here's one I prepared earlier" would be cute. I think other people's threads pointing out to scale globally it has to be something anyone can do, and right now Intel appear to have set the RAND conditions or whatever IPR they lock into this higher than other people want to pay.

Re: What makes Intel Optane stand out (2023)

#154

I wish it had made it to thumb drives for its superior data retention. It doesn't work by holding a charge; it's phase-change, so it retains data longer.

I encountered FRAM chips in one PCB I worked with. It's like flash, but holds data forever and does not wear. One could easily create usb drive of small capacity using these chips. They are a bit expensive, but that longevity could be a market differentiator. So tech is there if you think you can sell it.

Re: What makes Intel Optane stand out (2023)

#155

Earlier quoted context omitted.

You need CXL to extend the cache coherency properties of actual RAM over a remote link. That's costly tech. Otherwise, you're relying on the OS (and even the compiler/basic libraries, since you need to make fences, etc. OS-visible) to paper over the differences by doing its own implementation of distributed shared memory (this is known as a 'SSI' or single-system-image approach) which has significant challenges and i…

I didn't mean anything like that. Just the equivalent of a GPU with the ability to run arbitrary CPU oriented programs. Of course GPUs do many tasks very well but there are also plenty of problems that aren't well suited to them. Well I suppose I've answered my own question at this point. There probably just aren't enough real world problems that aren't amenable to running on a GPU while also being either compute or…

Things like that existed in the category of accelerator cards. Xeon Phi (Knights) is one example, focused on core count. Some from HP have soldered on SSDs too. You also had blade servers which is more focused on that use case, though that's going out of style.

I don't think PCIe is really a good fit for general CPU tasks. You need big heatsinks and power and can't fit that much RAM on board.

Re: What makes Intel Optane stand out (2023)

#156
post #118

Earlier quoted context omitted.

Isn't this addressed by newer PCIe standards? Of course, even the "new" Optane media reviewed in OP is stuck on PCIe 4.0...

No; the issue with the DIMMs wasn’t drivers. The issue was that the only people allowed to target the DIMMs directly were the xeon hardware team. There was a startup doing good work with similar storage chips that were pin (BGA) compatible with standard memory. Not sure what happened to them. That’d be easier to program than xpoint. As for the new PCIe standard (you probably mean CXL), that’s also basically dead on a…

Interesting perspective re CXL synchronous API. Wouldn't things like OOO execution and speculation help with that? And anyway the latency is supposed to be comparable to NUMA latency, is that really such a deal breaker?

Re: What makes Intel Optane stand out (2023)

#157
post #119
post #99

Earlier quoted context omitted.

I have tried multiple enterprise SSD's, for sync writes. Nothing comes close to Optane Dimm, even Optane NVMe is 10x slower than PDIMMS. https://forums.servethehome.com/index.php?threads/so-i-teste...

Yeah, but what percentage of the time was the CPU stuck blocked on the memory bus for the two workloads? Also, how was it guaranteeing the page writes were atomic?

You are correct that writes are not guaranteed to be atomic. PostgreSQL has a feature to handle torn pages: full_page_writes = on, which is enabled by default. This means you do double write.

For RocksDB I believe torn pages are not a problem as SST files are immutable. But correct me if I am wrong.

Re: What makes Intel Optane stand out (2023)

#158
post #129

Earlier quoted context omitted.

We don't care about (TBW/TB) at the consumer level, we care about (TBW/$), and 3D TLC was far, far cheaper per TB, so much so that TBW/$ was not a numerical advantage of Optane. That left ONLY the near-RAM-read-latency, which is only highly beneficial on specific workloads. Then they didn't invest in expanding killer app software that could utilize that latency, and didn't drop prices sufficiently to make it highly c…

In 2018, with Optane drives launching around $1.50/GB and TLC flash drives around $0.15/GB, it wasn't that much cheaper. As far as I'm aware Optane had a lot more than 10x the endurance.

10x endurance and 10x latency reduction, for 10x the price. It was closer to 5x the price with the closeout firesale after it was killed.

Re: What makes Intel Optane stand out (2023)

#160

My kingdom for a MicroSD card with Optane inside. My dashcam wants it soooo badly.

Its not Optane but there are SLC microSD cards!

I have a few of them, and yes, they're worth every penny!

(For posterity: Mouser and Digikey et alia carry these things and let you search/filter by cell type. They're not cheap.)

Post reply on HN