The Truth About Numonyx PCM: A Revolution That Hasn't Happened
Joke instead preface:
- You heard, Ivanov won the Volga lottery.
- Actually, not Ivanov, but Rabinovich, not Volga, but a hundred rubles, not in a lottery, but in preference, and did not win, but lost.
The other day, a link to a stunning news appeared on Habré: there was a revolution in memory technology, and soon we will have a single memory super-duper memory based on chalcogenide materials, using phase transitions caused by heating when electric current is passed through the cell. It works fast as DRAM, and even non-volatile, like Flash.
Unfortunately, almost everything written in the article is wrong ...
So that habrasoobshchestva not create a wrong idea about what is happening in the world of memory, I would like to tell the true story of PCM.
')
To a quite reasonable question from the gallery: “And who are you, so to speak, do we supposedly listen to your truth?” I can say that I work in the development department of a memory manufacturer (including PCM) and “keep my finger on the pulse” ".
Comment on the points of nonsense and factual errors on the link will not, they are too many. In one sentence: RSM is stillborn technology, there will be no revolution (and there was not).
The idea of ​​RSM is old (patent obtained in 1966), from a scientific point of view everything is clear and simple. Some materials (chalcogenide glasses, for example) have two quasistable states: crystalline and amorphous. In crystalline they conduct current, and in amorphous they do not. Consequently, by measuring the resistance of such a cell, information can be read.
How to write / erase it? Everything is quite simple: if the chalcogenide is heated to a high temperature (but below the melting point), it will turn into an amorphous state. If heated even more, it will melt, but upon subsequent cooling, it will recrystallize into the conducting phase.
About 7 years ago there was a certain research boom (no longer in the academy, but in the industry), due to the fact that there was a real opportunity to put the technology into mass production. Several companies licensed it, many conducted research. Then it seemed to many that what the PCM is about is the future.
About 3 years ago it became clear that nothing good would come of it: too expensive to produce, materials very inconvenient, problems with uneven recrystallization. If the cells are made small enough (and the large ones are not needed by anyone), then the resistances of the unsuccessfully recrystallized cell and its neighbors in the amorphous state will be almost the same. The main problem is the thermal insulation of the cells - it is difficult to hide from the heat in a small volume, and as soon as one cell is switched, the neighboring cells gradually lose their crystallinity.
At the same time there was a breakthrough in the flash memory market: multi-level cells appeared. When writing / reading, more than two signal levels can be distinguished from them, and, therefore, two, three, and in the future, four bits can be stored in one physical cell. In addition, many problems of miniaturization were solved: now flash memory, made according to the technical process of about 20 nm, is included in mass production.
However, a couple of companies (the first was Numonyx) released chips: just to minimize losses and not lose face. C 2008, you can easily buy PCM chips, but almost no orders. The recording density (Numonyx chips have a capacity of 128Mbit) is too low even for mobile applications. Samsung, with some delay, released 512Mbit chips, made by more advanced technology. But the queue of buyers is also not observed.
It is not planned to finish the technology (miniaturization below 60nm is almost impossible), for example, in our company, for example, 5 years ago there was a whole department that worked on PCM, today one person (rather stupid, he just had to stick it somewhere) . The only chance (in my opinion) is to “implant” a small amount of PCM into flash drive controllers (for caching), but DRAM is usually bypassed there.
Where do the legs of this home-grown sensation grow (and what’s the joke from the beginning of the article)?
Many sites have recently written about new PCM Numonyx chips. Basically, these are sets of common words about prospects, such as the one to which the source topic refers. They are written, for the most part simply by journalists "by the gland", who, unfortunately, are often not at all in the subject.
The news on the engineering resource sounds like "Numonyx announces a delay in the release of PCM memory of 1 Gbit . "
In fact, they cannot even repeat the result of Samsung ... But to maintain their reputation (Numoniks is now in a difficult situation, the former founders of STM and Intel are selling the company to Micron), marketers have thought up to renew the memory of two years ago with a new name and call a press conference from journalists “not in topic "to create an informational buzz about their success.
I am sure that in a few years everyone will calm down and stop releasing PCM, which, unfortunately, turned out to be a stillborn product.
- You heard, Ivanov won the Volga lottery.
- Actually, not Ivanov, but Rabinovich, not Volga, but a hundred rubles, not in a lottery, but in preference, and did not win, but lost.
The other day, a link to a stunning news appeared on Habré: there was a revolution in memory technology, and soon we will have a single memory super-duper memory based on chalcogenide materials, using phase transitions caused by heating when electric current is passed through the cell. It works fast as DRAM, and even non-volatile, like Flash.
Unfortunately, almost everything written in the article is wrong ...
So that habrasoobshchestva not create a wrong idea about what is happening in the world of memory, I would like to tell the true story of PCM.
')
To a quite reasonable question from the gallery: “And who are you, so to speak, do we supposedly listen to your truth?” I can say that I work in the development department of a memory manufacturer (including PCM) and “keep my finger on the pulse” ".
Comment on the points of nonsense and factual errors on the link will not, they are too many. In one sentence: RSM is stillborn technology, there will be no revolution (and there was not).
The idea of ​​RSM is old (patent obtained in 1966), from a scientific point of view everything is clear and simple. Some materials (chalcogenide glasses, for example) have two quasistable states: crystalline and amorphous. In crystalline they conduct current, and in amorphous they do not. Consequently, by measuring the resistance of such a cell, information can be read.
How to write / erase it? Everything is quite simple: if the chalcogenide is heated to a high temperature (but below the melting point), it will turn into an amorphous state. If heated even more, it will melt, but upon subsequent cooling, it will recrystallize into the conducting phase.
About 7 years ago there was a certain research boom (no longer in the academy, but in the industry), due to the fact that there was a real opportunity to put the technology into mass production. Several companies licensed it, many conducted research. Then it seemed to many that what the PCM is about is the future.
About 3 years ago it became clear that nothing good would come of it: too expensive to produce, materials very inconvenient, problems with uneven recrystallization. If the cells are made small enough (and the large ones are not needed by anyone), then the resistances of the unsuccessfully recrystallized cell and its neighbors in the amorphous state will be almost the same. The main problem is the thermal insulation of the cells - it is difficult to hide from the heat in a small volume, and as soon as one cell is switched, the neighboring cells gradually lose their crystallinity.
At the same time there was a breakthrough in the flash memory market: multi-level cells appeared. When writing / reading, more than two signal levels can be distinguished from them, and, therefore, two, three, and in the future, four bits can be stored in one physical cell. In addition, many problems of miniaturization were solved: now flash memory, made according to the technical process of about 20 nm, is included in mass production.
However, a couple of companies (the first was Numonyx) released chips: just to minimize losses and not lose face. C 2008, you can easily buy PCM chips, but almost no orders. The recording density (Numonyx chips have a capacity of 128Mbit) is too low even for mobile applications. Samsung, with some delay, released 512Mbit chips, made by more advanced technology. But the queue of buyers is also not observed.
It is not planned to finish the technology (miniaturization below 60nm is almost impossible), for example, in our company, for example, 5 years ago there was a whole department that worked on PCM, today one person (rather stupid, he just had to stick it somewhere) . The only chance (in my opinion) is to “implant” a small amount of PCM into flash drive controllers (for caching), but DRAM is usually bypassed there.
Where do the legs of this home-grown sensation grow (and what’s the joke from the beginning of the article)?
Many sites have recently written about new PCM Numonyx chips. Basically, these are sets of common words about prospects, such as the one to which the source topic refers. They are written, for the most part simply by journalists "by the gland", who, unfortunately, are often not at all in the subject.
The news on the engineering resource sounds like "Numonyx announces a delay in the release of PCM memory of 1 Gbit . "
In fact, they cannot even repeat the result of Samsung ... But to maintain their reputation (Numoniks is now in a difficult situation, the former founders of STM and Intel are selling the company to Micron), marketers have thought up to renew the memory of two years ago with a new name and call a press conference from journalists “not in topic "to create an informational buzz about their success.
I am sure that in a few years everyone will calm down and stop releasing PCM, which, unfortunately, turned out to be a stillborn product.
Source: https://habr.com/ru/post/92079/
All Articles