6 ms·

FreeBSD on the POWER8: it's alive

78 points by fcambus 12y ago

rwmj 12y ago
It's great to see IBM move slowly towards an open POWER architecture. At Red Hat we have access to lots of POWER7 and POWER8 hardware and it really is the fastest hardware available for anything, many times faster than x86-64 (of course with a massive cost and power budget to go with that). But when am I going to be able to buy POWER8 hardware for myself?? Is the expensive and experimental Tyan mobo the only thing?
- ghuntley 12y ago
  It's worth nothing that IBM /does/ have a virtual loaner program that provides free access to POWER7/POWER8 servers for software development. System i (OS/400), System P (AIX/Linux/FreeBSD [eventually]) http://www-304.ibm.com/partnerworld/wps/pub/systems/vlp http://www-304.ibm.com/partnerworld/wps/pub/systems/vlp Current spec is a 8286-42A @ http://www-03.ibm.com/systems/power/hardware/s824/specs.html http://www-03.ibm.com/systems/power/hardware/s824/specs.html
  - justincormack 12y ago
    Also if you dont want to sign up (it is not a lot of hassle, but you have to use a VPN and instances only stay up a week), OVH has a Power8 cloud with a small and big instance size. https://www.runabove.com/instances/ibm-power8.xml https://www.runabove.com/instances/ibm-power8.xml
  - scentoni 12y ago
    s/worth nothing/worth noting
- frozenport 12y ago
  >>is the fastest hardware available for anything, many times faster than x86-64 Thats a rather bold claim. Do you know of any benchmarks for this?
  - jacquesm 12y ago
    It's fairly common knowledge, not really a bold claim at all, but for your convenience I've linked a benchmark above. The IO (10+ PCIex slots) and CPU to memory bandwidth for a Power8 machine are pretty impressive. It's essentially a mainframe in a really small box.
- Gregordinary 12y ago
  Not 100% sure if it's POWER8, but there is a project to produce a modern 64-bit PowerPC notebook: http://www.powerpc-notebook.org/en/ http://www.powerpc-notebook.org/en/
  - wmf 12y ago
    I'm 100% sure Power8 won't fit into a notebook. Looks like they're using a Freescale network-oriented SoC.
    - yellowapple 12y ago
      Sure it could, hypothetically. It wouldn't be exactly fashionable or portable - think gargantuan Dell "portable workstations" instead of Macbook Airs - but it's hypothetically doable. And quite honestly, I'd be willing to sell my home and live in my car for the next decade or so in order to afford such a beauty should someone try and make one. Especially if it had some high-end GPUs wired directly into it via CAPI. Its battery life would be 4 seconds long, it would weigh more than a Ford truck, and it would run hot enough to cook my breakfast in the morning, but god damn would it be worth it.
jacquesm 12y ago
Interesting development, let's hope that IBM will pick this up and run with it as an officially supported OS at some point. It would give lots more room upwards for when you're about to max out a single server x86 set-up. It won't be long before someone runs a top-of-the-line X86 to Power8 benchmark using FreeBSD. Here is such a comparison using linux: http://www.itjungle.com/tfh/tfh092914-story02.html http://www.itjungle.com/tfh/tfh092914-story02.html
- tedunangst 12y ago
  Eh. IBM chip is faster at IBM selected benchmarks? Shocker. Also no mention of price.
  - cssmoo 12y ago
    Not terrible price: http://www-03.ibm.com/systems/power/hardware/s812l-s822l/browse.html http://www-03.ibm.com/systems/power/hardware/s812l-s822l/bro... Comparable to shonky-ass x64 HP kit. Perhaps I'm crazy but I'd love to see Windows Server on this as well. A non-x86 based platform alternative would be a refreshing change in that area.
  - tormeh 12y ago
    It does illustrate a point, though. I would guess Intel is faster overall, but for the right applications POWER and SPARC can be faster.
    - yellowapple 12y ago
      Namely, most applications that benefit from parallelism will benefit from POWER or SPARC, since both platforms pack a high number of cores and a high number of concurrent threads per core. This doesn't sound like much for consumer devices (PowerPC was the gold standard for gaming consoles last generation, but now Nintendo's the only one still using POWER in any significant application), but with Erlang/Elixir and Scala (among other languages renowned for facilitating "easy" parallelism) being the latest hot shit right now, these advantages are going to be much more significant.
tiffanyh 12y ago
Great use for Erlang (e.g. WhatsApp)? Given that Erlang can scale near linearly with more cores and it's not uncommon to have 100+ cores for a single Power8 node - it makes me wonder how much investigation has gone into Power8 as an architecture for companies who have highly network centric products like WhatsApp. Edit: WhatsApp is a highly network centric product and they also run on FreeBSD / Erlang. Edit 2: If you want to try out Power8, OVH has cloud compute available of Power8 at https://www.runabove.com/instances/ibm-power8.xml https://www.runabove.com/instances/ibm-power8.xml
- yellowapple 12y ago
  I think that scalability aspect is precisely why Google (IIRC) is on the OpenPOWER board; there's been hinting that they're interested in developing their own server boards in-house based on POWER8 (or perhaps a future generation).
bch 12y ago
- he's figured out the KVM issues, their lack of support for some mandated hypervisor APIs and other bugs - then found the existing powerpc pmap (physical memory management) code wasn't very SMP friendly - also found the PS3 hypervisor layer isn't thread-safe These are some of the reasons that I care about NetBSD (which runs on some 56 architectures with its single codebase) so much -- the real-world work supporting real-world architectures can pay off everywhere -- it's not just academic work or nostalgia/eccentricity.
- yellowapple 12y ago
  Being highly-portable also helps to make sure that your code has a maximum of cleanliness and correctness; this is one of the primary reasons why OpenBSD, for example, emphasizes the importance of being portable to everything from x86 and POWER systems to Zaurus PDAs and VAXen, even though portability isn't an explicit objective of OpenBSD like it is for NetBSD.
  - tiffanyh 12y ago
    I love NetBSD, don't get me wrong, but inevitably you will end up with a messing subsystem if managing multiple architectures because you won't be able to avoid situation like this: if (x86) // do ... else (Power) // do ... else (PPC) // do ...
    - bch 12y ago
      From http://www.netbsd.org/about/portability.html http://www.netbsd.org/about/portability.html: > Every effort is made to keep everything cleanly split into 'Machine Dependent' (MD) and 'Machine Independent' (MI) areas. For example, an Ethernet chipset would have a single MI core hardware driver, which would be matched with appropriate MD bus attachment code for a given platform. Not all drivers are as clean as we would like for historical reasons, but any new driver will be, and old drivers are in the process of being converted across. Machine dependent parts are built to a hardware abstraction layer (HAL)[1] interface, so that spec gives a single abstraction that the hardware-dependant code is built against, and can operate w/ the rest of the OS. As I understand it, this is facilitated by the _build system_ (not a mess of if/then in code) like this[2]: .if ${ARCHSUBDIR} == "mips64" ARCHSUBDIR= mips .endif .if ${ARCHSUBDIR} == "powerpc64" ARCHSUBDIR= powerpc .endif These are also the sorts of problems that are addressed by considering "lots and lots" of targets -- it'd be unmaintainable otherwise. [1]: http://en.wikipedia.org/wiki/NetBSD#Portability http://en.wikipedia.org/wiki/NetBSD#Portability [2]: /usr/src/sys/arch/Makefile EDIT: ws s/it's be/it'd be/ s/that spec give/that spec gives/
webaholic 12y ago
Can someone with access please run some cpu benchmarks ala anandtech? Thanks!
- erikarn 12y ago
  Sure, which ones would you like me to run? (adrian@freebsd.org)

78 points by fcambus 12y ago

rwmj 12y ago

It's great to see IBM move slowly towards an open POWER architecture. At Red Hat we have access to lots of POWER7 and POWER8 hardware and it really is the fastest hardware available for anything, many times faster than x86-64 (of course with a massive cost and power budget to go with that). But when am I going to be able to buy POWER8 hardware for myself?? Is the expensive and experimental Tyan mobo the only thing?

ghuntley 12y ago
It's worth nothing that IBM /does/ have a virtual loaner program that provides free access to POWER7/POWER8 servers for software development. System i (OS/400), System P (AIX/Linux/FreeBSD [eventually]) http://www-304.ibm.com/partnerworld/wps/pub/systems/vlp http://www-304.ibm.com/partnerworld/wps/pub/systems/vlp Current spec is a 8286-42A @ http://www-03.ibm.com/systems/power/hardware/s824/specs.html http://www-03.ibm.com/systems/power/hardware/s824/specs.html
- justincormack 12y ago
  Also if you dont want to sign up (it is not a lot of hassle, but you have to use a VPN and instances only stay up a week), OVH has a Power8 cloud with a small and big instance size. https://www.runabove.com/instances/ibm-power8.xml https://www.runabove.com/instances/ibm-power8.xml
- scentoni 12y ago
  s/worth nothing/worth noting
frozenport 12y ago
>>is the fastest hardware available for anything, many times faster than x86-64 Thats a rather bold claim. Do you know of any benchmarks for this?
- jacquesm 12y ago
  It's fairly common knowledge, not really a bold claim at all, but for your convenience I've linked a benchmark above. The IO (10+ PCIex slots) and CPU to memory bandwidth for a Power8 machine are pretty impressive. It's essentially a mainframe in a really small box.
Gregordinary 12y ago
Not 100% sure if it's POWER8, but there is a project to produce a modern 64-bit PowerPC notebook: http://www.powerpc-notebook.org/en/ http://www.powerpc-notebook.org/en/
- wmf 12y ago
  I'm 100% sure Power8 won't fit into a notebook. Looks like they're using a Freescale network-oriented SoC.
  - yellowapple 12y ago
    Sure it could, hypothetically. It wouldn't be exactly fashionable or portable - think gargantuan Dell "portable workstations" instead of Macbook Airs - but it's hypothetically doable. And quite honestly, I'd be willing to sell my home and live in my car for the next decade or so in order to afford such a beauty should someone try and make one. Especially if it had some high-end GPUs wired directly into it via CAPI. Its battery life would be 4 seconds long, it would weigh more than a Ford truck, and it would run hot enough to cook my breakfast in the morning, but god damn would it be worth it.

jacquesm 12y ago

Interesting development, let's hope that IBM will pick this up and run with it as an officially supported OS at some point. It would give lots more room upwards for when you're about to max out a single server x86 set-up. It won't be long before someone runs a top-of-the-line X86 to Power8 benchmark using FreeBSD. Here is such a comparison using linux: http://www.itjungle.com/tfh/tfh092914-story02.html http://www.itjungle.com/tfh/tfh092914-story02.html

tedunangst 12y ago
Eh. IBM chip is faster at IBM selected benchmarks? Shocker. Also no mention of price.
- cssmoo 12y ago
  Not terrible price: http://www-03.ibm.com/systems/power/hardware/s812l-s822l/browse.html http://www-03.ibm.com/systems/power/hardware/s812l-s822l/bro... Comparable to shonky-ass x64 HP kit. Perhaps I'm crazy but I'd love to see Windows Server on this as well. A non-x86 based platform alternative would be a refreshing change in that area.
- tormeh 12y ago
  It does illustrate a point, though. I would guess Intel is faster overall, but for the right applications POWER and SPARC can be faster.
  - yellowapple 12y ago
    Namely, most applications that benefit from parallelism will benefit from POWER or SPARC, since both platforms pack a high number of cores and a high number of concurrent threads per core. This doesn't sound like much for consumer devices (PowerPC was the gold standard for gaming consoles last generation, but now Nintendo's the only one still using POWER in any significant application), but with Erlang/Elixir and Scala (among other languages renowned for facilitating "easy" parallelism) being the latest hot shit right now, these advantages are going to be much more significant.

tiffanyh 12y ago

Great use for Erlang (e.g. WhatsApp)? Given that Erlang can scale near linearly with more cores and it's not uncommon to have 100+ cores for a single Power8 node - it makes me wonder how much investigation has gone into Power8 as an architecture for companies who have highly network centric products like WhatsApp. Edit: WhatsApp is a highly network centric product and they also run on FreeBSD / Erlang. Edit 2: If you want to try out Power8, OVH has cloud compute available of Power8 at https://www.runabove.com/instances/ibm-power8.xml https://www.runabove.com/instances/ibm-power8.xml

yellowapple 12y ago
I think that scalability aspect is precisely why Google (IIRC) is on the OpenPOWER board; there's been hinting that they're interested in developing their own server boards in-house based on POWER8 (or perhaps a future generation).

bch 12y ago

- he's figured out the KVM issues, their lack of support for some mandated hypervisor APIs and other bugs - then found the existing powerpc pmap (physical memory management) code wasn't very SMP friendly - also found the PS3 hypervisor layer isn't thread-safe These are some of the reasons that I care about NetBSD (which runs on some 56 architectures with its single codebase) so much -- the real-world work supporting real-world architectures can pay off everywhere -- it's not just academic work or nostalgia/eccentricity.

yellowapple 12y ago
Being highly-portable also helps to make sure that your code has a maximum of cleanliness and correctness; this is one of the primary reasons why OpenBSD, for example, emphasizes the importance of being portable to everything from x86 and POWER systems to Zaurus PDAs and VAXen, even though portability isn't an explicit objective of OpenBSD like it is for NetBSD.
- tiffanyh 12y ago
  I love NetBSD, don't get me wrong, but inevitably you will end up with a messing subsystem if managing multiple architectures because you won't be able to avoid situation like this: if (x86) // do ... else (Power) // do ... else (PPC) // do ...
  - bch 12y ago
    From http://www.netbsd.org/about/portability.html http://www.netbsd.org/about/portability.html: > Every effort is made to keep everything cleanly split into 'Machine Dependent' (MD) and 'Machine Independent' (MI) areas. For example, an Ethernet chipset would have a single MI core hardware driver, which would be matched with appropriate MD bus attachment code for a given platform. Not all drivers are as clean as we would like for historical reasons, but any new driver will be, and old drivers are in the process of being converted across. Machine dependent parts are built to a hardware abstraction layer (HAL)[1] interface, so that spec gives a single abstraction that the hardware-dependant code is built against, and can operate w/ the rest of the OS. As I understand it, this is facilitated by the _build system_ (not a mess of if/then in code) like this[2]: .if ${ARCHSUBDIR} == "mips64" ARCHSUBDIR= mips .endif .if ${ARCHSUBDIR} == "powerpc64" ARCHSUBDIR= powerpc .endif These are also the sorts of problems that are addressed by considering "lots and lots" of targets -- it'd be unmaintainable otherwise. [1]: http://en.wikipedia.org/wiki/NetBSD#Portability http://en.wikipedia.org/wiki/NetBSD#Portability [2]: /usr/src/sys/arch/Makefile EDIT: ws s/it's be/it'd be/ s/that spec give/that spec gives/

webaholic 12y ago

Can someone with access please run some cpu benchmarks ala anandtech? Thanks!

erikarn 12y ago
Sure, which ones would you like me to run? (adrian@freebsd.org)