reliabilityguy

I am confused. Are you ashamed of “our” crimes (I’m not who the “our” refers to), and this is why you are fine with the atrocities done by Houthis?

> Just because something can be done doesn't make it good for the society.

In this case the consequences will make people to elect a better government next time and try better ideas. Democracy at work.

> That goes well beyond simply making sure that people are able to care for themselves.

I am not sure in understand the issue, if Japanese want this, what’s the problem?

> ACA?

How’s ACA hard-hitting on corporations?

The 15/20 rules is what resulted in recorder profits in health insurance business, how’s this anti-corporate?

HBM is a stack of DRAMs, so there is no “instead”.

> They're still recording everywhere and sending a transcription to iCloud and they're doing it without consent.

Plenty of jurisdictions are single consent.

Moreover, what is the difference with using your phone to record?

> when it is in fact a well-documented and pretty mundane practice

Practiced by Apple? Just because someone is doing it doesn’t mean that every company does it.

I am not the one making claims here.

You can’t claim A, and then when someone asks you for a proof reply with “prove that it’s not A”.

There is no factual accuracy at all, only speculation and empty claims.

Do you have a proof that Apple has bot farms to push its narrative?

> load all the LLM parameters or KV cache in RAM and exclusively let it perform GEMV and let it rip.

Won’t you have a bunch of extra reads/writes via the CPU because these DIMMs won’t be able to compute matrix multiplications?

Rollie calculation, like the online softmax in FA, implies a centralized computing unit that does the compute and stores the intermediate results in its registers. With PIM you have no centralized compute unit, you have a bunch of memory, and a bunch of MACs all over the place.

How would you do map-reduce across multiple DIMMs w/o extra reads/writes?

PIM implies some sort of distributed compute, which can work for some cases, but I am not sure LLMs are one of them.

How exactly the map-reduce will happen though? Won’t you need to do it host-side, or make a lot of reads and writes?

Also, doesn’t it mean that you forgo batching?

> You could run MACs directly in RAM

Sure, MACs are nice. However, unless there other, PIM-specific/optimal, algorithms, regular matrix multiplication algorithms like tiling-based won’t work here I think — how would the tile be shared? By doing read/write all the time?

Interesting that Samsung still pursues PIM. IIRC they had a paper in ISCA21 or 22 where they showed HBM2 module with PIM, which back then impressed me quite a lot.

That being said, I am not sure what’s the killer application for this technology, and without such application adoption is unlikely.

I am not sure a vote of non-confidence would have helped here — there are only two parties.

Can you elaborate what you mean?

I am not sure we are arguing here on the merit of the argument itself. You are fine with China’s genocide TODAY because China is the “lesser of two evils”?

I do not understand this. It sounds like a cultist/religious position.

No, I mean licenses to use the ISA. The act of paying for the license has nothing to with Apple being or not being one of the founders.

> Sure, but not the way West conducted it, by going and killing generations of people.

Ridiculous statement. China is doing it today — uighurs are actively erased bit by bit.

> Look at India, before British colonization they accounted for almost 25% GDP of the world (~1700s), now it is around 3-4%

This statement is rooted in “zero sum” logic, and based on the assumption that without colonization India should have stayed with its 25% or more share. Hypotheticals are hypotheticals, and nothing more.

> You keep defending the West,

I’m not defending anyone. I find it ridiculous that you are so high on “the west is bad, China good” horse and completely oblivious to so many things that China does, which are objectively worse.

China is imperialist as any other country. They extract resources from others like the British 100 years ago.

This is hilarious that this whole “west is bad” makes people blind.

> Imperialism costed millions of lives

Chinese policies resulted in 10s of millions of their own people dead, and many millions in labor camps today.

Sure, “west is bad” but China “has its problems”.

> I'd call it a whataboutism

I would say that saying about anything “whataboutism” is a cheap way out.

China does industrial espionage for the same end goals as American companies work on new things: market domination. So, I find it funny when someone is mad about American companies but not Chinese, while both pursuit the same goals.

> all for profit companies are greedy by their nature, but US companies are on a different level because you can 'buy' the government by lobbying or supporting their election campaigns

How would you call decades of chinese state-sponsored industrial espionage of western IP? Benevolent act?

> I don't care about internal politics of a state, as long as they are not killing people deliberately

What about forced cultural assimilation and re-education camps? Or, as long as it’s confined to the borders of a country doing so, it is fine?

> How government is structured is up to Chinese people to decide

Funny take given the events of 1989

> I wonder just how different their processes really are

Night and day.

Apples has in-house silicon dev groups starting (at least) 2008 w/ acquisition of PA semi for their low power CPUs, followed by the acquisition of Anobit in 2012(?). Apple does their own architecture, microarchitecture, etc, full cycle development.

Apple licenses ISA, not designs.

> you're describing a system in which a refresh causes so much disturbance that another refresh is immediately required.

Well, if you need to refresh a bunch of rows after any other read, you will spend more time refreshing things rather doing useful work.

They are, that’s why ECC exists, and those who are willing to pay for it, can get it.

> It's my long standing opinion[1] that we shouldn't accept any RAM that can be subject to random bit flips.

Every system fails eventually.

> Okay then refresh the whole thing and not just a single row whenever TRR is triggered.

How would you refresh the whole DRAM? Refresh is simply reading the row and writing it back, it is sequential in nature.

> Or interleave activations with refreshes. Say, after every 5 activations, refresh the next row in the refresh cycle.

It makes zero sense. Refresh is disturbance in itself. You are entering a recursion here where the mere act of refreshing increases the counter values of victim rows making refresh even more frequent. At the end you have DRAM that you cannot read from or write to at all because it’s always refreshes itself.

> I'm not a DRAM expert

Yes

Interesting read. To his defense: the argument is not that Optionals are too complicated, but that it’s a wrong path for language design — instead programmers have to test their code properly.

I like optionals, but I see his point too.

> Doing both at once is much harder.

I think with sequential approach (translate -> make idiomatic) it is easier. However, I am not sure that monetary difference is going to be as stark as $165k vs 3 engineers/year. Especially, if you consider that no one knows that is what in the code at the end.

Sure, you can argue that now it doesn’t matter — agents and all that, but I am not so sure.