This is funny, because in 100 years, computers with 100 yottabytes of memory will be standard, and they'd be laughing at a headline like this one.
HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
41–50 of 57 posts
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#42Wasn't the Machine supposed to be a memristor computer? Did they ditch the project?
Yes, but I think the idea was that The Machine was a constellation of technology projects for memory-focused computers. Although the memristor portion did not live up to their claimed expectations (unfortunate but unsurprising), the overall project of a memory-focused computer continues under the brand of The Machine.
(My outside-in interpretation. Maybe someone from HP can add more concrete detail and insight.)
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#43I am not sure if I should be impressed until they reveal more details. 160 TB over 40 nodes is 4 TB per node, which assuming 512 GB dimms requires only 8 dimms (4 per socket) or with 256 GB dimms 16 dimms (8 per socket). How fast is the interconnect? 100Gb/s? So far it sounds like an common high memory HPC cluster unit with really unknown technical parameters.
Where can you get 512 GB dimms?
I think even 128GB RDIMMs using TSV and 8Gbit DDR4 is expensive at this point. Interestingly, I have not even seen any 16Gbit DDR4 announcements.
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#44Wasn't the Machine supposed to be a memristor computer? Did they ditch the project?
I recall having read that it was not yet economical to mass produce memristors, which had to compete against modern transistor chips.
So they went ahead with the unified memory architecture, just without the memristor component.
One could argue that this architecture is future-ready, once memristors become economical to produce
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#45How does the Machine recover from software errors? If it has only persistent RAM, then when some important program goes bonkers (it always goes), there is no luxury of pressing the reset button to get back to a pristine state.
My guess is that there would be a microkernel service that is the core OS, like the BIOS, and it could be instructed to actively wipe the memory and reload the boot files when the machine freezes
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#46> The new prototype has 160 TB of shared memory spread across 40 physical nodes, interconnected using a high-performance fabric protocol. So basically something similar to RDMA scaled to 40 nodes, 4 TB RAM per node.
I think the interesting part is the programming model and not the amount of RAM. This computer has no separate storage area. If you want to manipulate data, you can do it in-place instead of reading it into a RAM-like working memory and then writing the result to a disk-like storage.
I think the interesting part is the programming model and not the amount of RAM. This computer has no separate storage area.
You can already mount a filesytem backed by ram, or mmap files that exist on disk. What existing distinction between RAM and storage is holding anything back?
I see a reason some techniques will get more popular, but I don't really seen the difference. Saving a load operation? Not having to worry about flushing to long term storage? Am I missing something?
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#47I am not sure if I should be impressed until they reveal more details. 160 TB over 40 nodes is 4 TB per node, which assuming 512 GB dimms requires only 8 dimms (4 per socket) or with 256 GB dimms 16 dimms (8 per socket). How fast is the interconnect? 100Gb/s? So far it sounds like an common high memory HPC cluster unit with really unknown technical parameters.
Isn't the idea that a single process can address all 160TB of memory as if it were local?
Of course you can cobble together that much memory when you're programming a distributed system of communicating processes. The interesting part would be programming it as if for one computer.
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#48Earlier quoted context omitted.
This comment currently has a reply ( http://i.imgur.com/N7BZJfu.png ) which shows why I wanted to leave my original post as a "riddle" rather than coming out and saying it. The following "crap" is not relevant to anything, only read it if you want the answer to why 1.6 KB is insufficient for a certain science fiction fantasy which will never, not in ten, a hundred, or a thousand years, become relevant any more than m…
You made a math error: You can encode about 300 addresses, at 36+8 = 44 bits per item, not 45 addresses, at 36*8 = 288 bits per address. The bits are additive, because we're storing an address and data, not a byte of data per bit of address.
thanks, corrected.
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#49Earlier quoted context omitted.
It's a 40 node mesh. Isn't that scale-out?
Let me rephrase rather. a scale out architecture with commodity machines (using Hadoop, kube, mesos etc.) versus an engineered system in hardware. A client purchased a an HP super dome (another highly engineered system) the cost per CPU core, gb ram, tb of storage was horrendous. More than 10x comparable commodity hardware solutions. 8 months later and they hit performance problems, space issues and adding capacity i…
This is a scale-out solution, it's just higher performance / lower effort than building your own solution from commodity harwdare. Enterprises often want to buy canned, supported solutions rather than expend the R&D to build their own.
What workload does this configuration make sense for? I don't know, especially with those ARM cores. My guess would be the cluster performs well on workloads similar to Apache Spark. But I don't know why you would choose one over the other.
Re: HPE unveils The Machine, a single-memory computer with 160 terabytes of memory
#50How does the Machine recover from software errors? If it has only persistent RAM, then when some important program goes bonkers (it always goes), there is no luxury of pressing the reset button to get back to a pristine state.
My guess is that there would be a microkernel service that is the core OS, like the BIOS, and it could be instructed to actively wipe the memory and reload the boot files when the machine freezes
There can be bug in a microkernel service (and as it is distributed, it is actually guaranteed it will crash unexpectedly at some point)