Monday, October 26, 2015

Extreme Tech reports on THE MACHINE

I earlier (June 13, 2014) reported on THE MACHINE as described by HP at the DISCOVER conference June 11, 2014.  I did not pass along the Sebastian Anthony comments at the time, but this will serve as good review of what was shared that day, and how it was received:

HP bets it all on The Machine, a new computer architecture based on memristors and silicon photonics, ·       By Sebastian Anthony on June 11, 2014 at 11:30 am

HP, one of the original 800lb Silicon Valley gorillas that has seen much happier days, is staking everything on a brand new computer architecture that it calls… The Machine. Judging by an early report from Bloomberg Businessweek, up to 75% of HP’s once fairly illustrious R&D division — HP Labs — are working on The Machine. As you would expect, details of what will actually make The Machine a unique proposition are hard to come by, but it sounds like HP’s groundbreaking work on memristors (pictured top) and silicon photonics will play a key role.
In the words of HP Labs, The Machine will be a complete replacement for current computer system architectures. There will be a new operating system, a new type of memory (memristors), and super-fast buses/peripheral interconnects (photonics). Speaking to Bloomberg, HP says it will commercialize The Machine within a few years, “or fall on its face trying.”
First things first, we’re probably not talking about a consumer computing architecture here, though it’s possible that technologies commercialized by The Machine will percolate down to desktops and laptops. Basically, HP used to be a huge player in the workstation and server markets, with its own operating system and hardware architecture, much like Sun. Over the last 10 years though, Intel’s x86 architecture has rapidly taken over, to the point where HP (and Dell and IBM) are essentially just OEM resellers of commodity x86 servers. This has driven down enterprise profit margins — and when combined with its huge stake in the diminishing PC market, you can see why HP is rather nervous about the future. The Machine, and IBM’s OpenPower initiative, are both attempts to get out from underneath Intel’s x86 monopoly.
HP started work on The Machine two years ago, when Martin Fink became CTO and head of HP Labs. He took a look at the components that HP Labs was working on — memristors, silicon photonics — and six months later he decided that it was time to pitch The Machine to HP CEO Meg Whitman. At the presentation, Fink said he expected 75% of HP Labs personnel to be dedicated to work on The Machine. Seemingly, Whitman agreed to the plan, because here we are in 2014 and HP is apparently staking its future on it.
While exact details are hard to come by, it seems The Machine is predicated on the idea that current RAM, storage, and interconnect technology can’t keep up with modern Big Data processing requirements. HP is working on two technologies that could solve both problems: Memristors could replace RAM and long-term flash storage, and silicon photonics could provide faster on- and off-motherboard buses. Memristors essentially combine the benefits of DRAM and flash storage in a single, hyper-fast, super-dense package. Silicon photonics is all about reducing optical transmission and reception to a scale that can be integrated into silicon chips (moving from electrical to optical would allow for much higher data rates and lower power consumption). Both technologies can be built using conventional fabrication techniques.
It sounds like The Machine would do away with RAM and external storage, instead packing tons of high-density memristor chips onto the motherboard, close to the CPU. (No word on what architecture the CPU might use, incidentally.) Everything would be connected together at high speed using silicon photonics — and you will no doubt be able to connect multiple Machines together via an optical interconnect, too.
The result would undoubtedly be a very fast device that opens up new ways of processing data. But, in all honesty, it isn’t hard to conceive of a fantabulous computer architecture that has monstrous processing power — but it is hard to actually build such a system in reality. The key element here will be software — both the operating system, which will have to be designed from the ground up for these new computing and storage paradigms, and the app ecosystem. The entire PC and server industry is already entrenched with x86, Windows, and Linux. It would require a monumental effort on HP’s behalf to create and support a brand new architecture that has almost no similarity to the platforms that developers have been targeting for the last 50 years.
But, I guess, when the only other option is selling PCs at 5% profit, and then padding the bottom line with overpriced printer ink, you sometimes have to make a Hail Mary pass — and that’s exactly what The Machine is.

No comments: