HP Enterprise shows off prototype of new computing architecture called "The Machine"

William Gayde

Posts: 382   +5
Staff

Hewlett Packard Enterprise last week unveiled a working prototype of their long awaited new computing architecture called "The Machine". Rather than focusing solely on a faster processor or more memory, The Machine employs a new memory-driven computing design.

HPE announced the project back in 2014 with up to 75% of their research labs focused on its development. The core idea behind memory-driven computing is using memory at the center of the design rather than the processor. Traditionally high performance computers used multiple processors each with their own set of memory. This creates bottlenecks and requires extra data to be transferred between these collections of memory. The Machine uses a single massive pool of memory across all the processors. 

The prototype system currently has 8 terabytes of memory but HP is looking to expand that to hundreds of terabytes in the future as more nodes are linked together. HP's goal is to use NVM (non-volatile memory) as a replacement for traditional DRAM. Instead of using copper interconnects on the board, HP has also developed a new X1 photonics module that uses light to transfer data in place of electrons. It is capable of transmitting up to 1.2Tbps at short distances.

The project has hit some bumps along the way however. The original design was slated to be released as a standalone product, but that idea has since changed. The Machine's core architecture will instead be built into future HP products. Elements like the photonics data transmission module and memristor technology -- memory that can retain data when powered down -- will make their way onto the market in 2018 or 2019.

The goal is to use this memory-focused technology to analyze huge data sets very quickly, for example to scan millions of financial transactions each second for fraud. HP claims The Machine will be able to address any data in a 160PB storage space in 250ns. As technology continues to evolve, these kinds of computations on the Petabyte and Exabyte scale don't look too far off.

Permalink to story.

 
Well, we have certainly reached a milestone in the computer world. We have gotten over the concept of making them smaller and thanks to HP, some of these processing boards are the size of a small surf board! If it ever makes into the residential world we'll need to build on an extra room or kick the kids out earlier in order to make room for it. Now we're waiting to see the peripherals!
 
It is capable of transmitting up to 1.2Tbps at short distances.
For a photonic platform this is not impressive at all. That's only 2x the speed of PCI Express 4.0.

I hope they actually got something more inventive than the product's name.
 
Last edited:
This gonna be exactly like HP's "revolutionary" memresistor that was revealed in what 2006 and never came out?
 
It is capable of transmitting up to 1.2Tbps at short distances.
For a photonic platform this is not impressive at all. That's only 2x the speed of PCI Express 4.0.

I hope they actually got something more inventive than the product's name.

Except that PCIe 4.0 hasn't been released yet. Heck, we still have boards out there that are only capable of PCIe 2.0 speeds, & unless the number of PCIe lanes is limited artificially by the board manufacturer we still don't have systems that will stress a PCIe 3.0 bus.
 
It is capable of transmitting up to 1.2Tbps at short distances.
For a photonic platform this is not impressive at all. That's only 2x the speed of PCI Express 4.0.

I hope they actually got something more inventive than the product's name.

Except that PCIe 4.0 hasn't been released yet. Heck, we still have boards out there that are only capable of PCIe 2.0 speeds, & unless the number of PCIe lanes is limited artificially by the board manufacturer we still don't have systems that will stress a PCIe 3.0 bus.
My thoughts exactly. Just because it's capable doesn't mean we're taking advantage of all available bandwidth.
 
HPE is virtually non-existant.
They have no product line and no market share. They periodically fish for customers for an idea, which is what this article is doing. As a business, HPE has no reason to exist, aside from selling off assets in order to show an income. Like Sears, HPE will fade into obscurity. They missed the mobile revolution and are no longer relevant to tech.
 
This gonna be exactly like HP's "revolutionary" memresistor that was revealed in what 2006 and never came out?

HP announced in 2006 or whenever that their labs had proven that memristors -- theoretical up to that point -- were now a real thing. The memristor was first described by Leon Chua of Berkeley in 1971. This was significant only because the memristor is a fundamental circuit component. Like resistors, inductors, and capacitors. That fundamental. But no one has actualy made one, much less figured out what to do with it.

HP was doing low-level research and built practical memristors. If they have real high density memories even just in the lab by now, that's pretty fast progress. Maybe this seems long because consumer product cycles are six months long these days and practically no one does basic research. But in terms of real science, not so long.
 
Back