Logs: liberachat/#haskell
| 2021-07-26 06:20:03 | <Axman6> | OpenMPI takes care of several things: it distributes binaries across clusters (it's almost exclusively used for HPC/scientific computing as far as I understand it), and abstracting network comminications between distinct executions - it's niche is things like large matrix computations where each node in the cluster computes a sub computation and then MPI helps to efficiently move the temporary results around and combine them into the full result |
| 2021-07-26 06:20:43 | <Axman6> | Distributed computing only makes sense if there's a network, so that's why I metnion it |
| 2021-07-26 06:21:04 | × | fawful quits (~guy@c-76-104-217-93.hsd1.wa.comcast.net) (Quit: WeeChat 3.2) |
| 2021-07-26 06:21:05 | <euouae> | Ok but I'm assuming they're not communicating with TCP, or are they? |
| 2021-07-26 06:21:50 | <Axman6> | if you're talking about MPI, that would depend on the cluster. MPI usually knows how to use things like infiniband, which is basically a networked PCI bus with direct memory access |
| 2021-07-26 06:22:14 | <Axman6> | but there's no reason not to use TCP either, it'll just be hiugher latency, which might be fine |
| 2021-07-26 06:22:46 | <euouae> | I guess you're right, adding seconds to a computation is not a big deal if it takes an hour or more |
| 2021-07-26 06:22:48 | <Axman6> | you did specifically ask about distributed computing, so if that's not what you axtually meant, I can give other answers :P |
| 2021-07-26 06:23:04 | <Axman6> | actually* |
| 2021-07-26 06:23:31 | <euouae> | I don't feel very comfortable in these topics as I've never done them before, so my apology if I'm flimsy hehe |
| 2021-07-26 06:23:42 | <euouae> | I've only had experience with parallelism within a single node |
| 2021-07-26 06:23:59 | × | thyriaen quits (~thyriaen@x4dbf5c5c.dyn.telefonica.de) (Remote host closed the connection) |
| 2021-07-26 06:24:19 | <euouae> | I think as a first introduction I might go with C++ and OpenMPI and switch to Haskell later if I can |
| 2021-07-26 06:24:22 | <Axman6> | Well we have lots of tools for that, and that book is basically the bible on the topic, do get reading! it's extremely approachable if you have the basics of Haskell down |
| 2021-07-26 06:24:34 | <euouae> | Oh so it does mention distributed computing as well? |
| 2021-07-26 06:24:45 | <Axman6> | if you're working on a single machine, OpenMPI is unnecessary. OpenMP will give you parallelism though |
| 2021-07-26 06:24:58 | <euouae> | I have access to a supercoputer |
| 2021-07-26 06:25:01 | <Axman6> | you don't seem to be talking about distributed computing |
| 2021-07-26 06:25:24 | <Axman6> | OpenMPI doesn't magically take a single binary and make it run across a whole cluster though |
| 2021-07-26 06:25:36 | <euouae> | I've never used OpenMPI |
| 2021-07-26 06:25:46 | <euouae> | It's on my todo list |
| 2021-07-26 06:28:34 | × | dunkeln quits (~dunkeln@94.129.69.87) (Ping timeout: 265 seconds) |
| 2021-07-26 06:30:50 | → | anandprabhu joins (~anandprab@94.202.243.198) |
| 2021-07-26 06:32:28 | → | dunkeln joins (~dunkeln@94.129.69.87) |
| 2021-07-26 06:34:29 | × | eggplantade quits (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) (Remote host closed the connection) |
| 2021-07-26 06:41:44 | → | chele joins (~chele@user/chele) |
| 2021-07-26 06:42:48 | → | mikoto-chan joins (~mikoto-ch@ip-193-121-10-50.dsl.scarlet.be) |
| 2021-07-26 06:46:18 | × | tommd quits (~tommd@75-164-130-101.ptld.qwest.net) (Ping timeout: 240 seconds) |
| 2021-07-26 06:46:21 | → | roboguy_ joins (~roboguy_@2605:a601:ac0e:ae00:6549:5b6:6ac5:3a5d) |
| 2021-07-26 06:46:54 | × | dunkeln quits (~dunkeln@94.129.69.87) (Ping timeout: 256 seconds) |
| 2021-07-26 06:47:36 | → | jco joins (~jco@213-64-226-9-no542.tbcn.telia.com) |
| 2021-07-26 06:48:48 | → | eggplantade joins (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) |
| 2021-07-26 06:53:22 | × | eggplantade quits (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) (Ping timeout: 258 seconds) |
| 2021-07-26 06:53:32 | × | Sgeo quits (~Sgeo@user/sgeo) (Read error: Connection reset by peer) |
| 2021-07-26 06:53:52 | → | lortabac joins (~lortabac@2a01:e0a:541:b8f0:a10:7118:963d:8400) |
| 2021-07-26 06:54:03 | × | roboguy_ quits (~roboguy_@2605:a601:ac0e:ae00:6549:5b6:6ac5:3a5d) () |
| 2021-07-26 06:54:37 | → | takuan joins (~takuan@178-116-218-225.access.telenet.be) |
| 2021-07-26 06:58:25 | → | ubert joins (~Thunderbi@77.119.200.110.wireless.dyn.drei.com) |
| 2021-07-26 06:59:10 | → | fendor joins (~fendor@178.165.161.47.wireless.dyn.drei.com) |
| 2021-07-26 07:05:02 | → | eggplantade joins (~Eggplanta@108-201-191-115.lightspeed.sntcca.sbcglobal.net) |
| 2021-07-26 07:07:26 | × | zmt01 quits (~zmt00@user/zmt00) (Ping timeout: 252 seconds) |
| 2021-07-26 07:07:38 | → | yoneda joins (~mike@193.206.102.122) |
| 2021-07-26 07:09:06 | → | jneira joins (~jneira@212.8.115.226) |
| 2021-07-26 07:09:20 | → | _ht joins (~quassel@82-169-194-8.biz.kpn.net) |
| 2021-07-26 07:10:08 | × | Erutuon quits (~Erutuon@user/erutuon) (Ping timeout: 256 seconds) |
| 2021-07-26 07:11:29 | → | derelict joins (~derelict@user/derelict) |
| 2021-07-26 07:12:52 | → | Morrow joins (~Morrow@147.161.8.32) |
| 2021-07-26 07:12:53 | → | PinealGlandOptic joins (~PinealGla@37.115.210.35) |
| 2021-07-26 07:14:44 | → | oxide joins (~lambda@user/oxide) |
| 2021-07-26 07:17:20 | × | wei2912 quits (~wei2912@112.199.250.21) (Quit: Lost terminal) |
| 2021-07-26 07:20:26 | × | slowButPresent quits (~slowButPr@user/slowbutpresent) (Quit: leaving) |
| 2021-07-26 07:26:34 | × | fossdd quits (~fossdd@sourcehut/user/fossdd) (Ping timeout: 256 seconds) |
| 2021-07-26 07:26:53 | → | fossdd joins (~fossdd@sourcehut/user/fossdd) |
| 2021-07-26 07:27:37 | × | m1dnight quits (~christoph@188.ip-51-91-158.eu) (Quit: WeeChat 3.1) |
| 2021-07-26 07:27:56 | → | m1dnight joins (~christoph@188.ip-51-91-158.eu) |
| 2021-07-26 07:28:22 | <juri_> | I haven't seen much haskell distributed computing stuff that 'just works', sadly. |
| 2021-07-26 07:28:27 | × | aplainzetakind quits (~johndoe@captainludd.powered.by.lunarbnc.net) (Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net) |
| 2021-07-26 07:28:37 | → | aplainzetakind joins (~johndoe@captainludd.powered.by.lunarbnc.net) |
| 2021-07-26 07:29:01 | <euouae> | juri_: so it's a thing for experts? |
| 2021-07-26 07:29:15 | <euouae> | I'm just getting into it myself, I'm hopeful |
| 2021-07-26 07:29:23 | <euouae> | at least I'm hopeful I can get it done in C++ |
| 2021-07-26 07:29:33 | <juri_> | I think it's just a thing that stalls every now and then, because there's not much demand. |
| 2021-07-26 07:29:39 | → | jushur joins (~human@user/jushur) |
| 2021-07-26 07:29:59 | <euouae> | It surprises me that scientific computing isn't bigger with Haskell |
| 2021-07-26 07:30:05 | <euouae> | It is very convenient for it |
| 2021-07-26 07:30:21 | <juri_> | I think one thing that has stuck in that area is accelerate. |
| 2021-07-26 07:30:30 | <juri_> | or data parallel haskell.. |
| 2021-07-26 07:31:19 | <juri_> | accelerate is neat, because if you write code within it's constraints, you can use a GPU and CPU at the same time.. |
| 2021-07-26 07:31:44 | <euouae> | Right but can you use multiple nodes? |
| 2021-07-26 07:31:52 | <juri_> | I'm just not much of a fan, because it requires real nvidia drivers. |
| 2021-07-26 07:32:05 | <euouae> | Real drivers? |
| 2021-07-26 07:32:26 | <juri_> | no, for that, you'd want cloud haskell. which seems to be stalling out. |
| 2021-07-26 07:32:39 | × | tzh quits (~tzh@c-24-21-73-154.hsd1.wa.comcast.net) (Quit: zzz) |
| 2021-07-26 07:33:41 | <Athas> | DPH is also long dead, but Accelerate is active. |
| 2021-07-26 07:34:29 | → | zeenk joins (~zeenk@2a02:2f04:a008:d600:18f2:3421:bac6:8f38) |
| 2021-07-26 07:35:08 | × | vysn quits (~vysn@user/vysn) (Remote host closed the connection) |
| 2021-07-26 07:36:12 | → | Obo joins (~roberto@70.pool90-171-81.dynamic.orange.es) |
| 2021-07-26 07:40:41 | → | gehmehgeh joins (~user@user/gehmehgeh) |
| 2021-07-26 07:41:25 | × | dagit quits (~dagit@2601:1c2:1b7f:9fa0:54b1:3520:a1d6:48b1) (Remote host closed the connection) |
| 2021-07-26 07:41:41 | → | dagit joins (~dagit@2601:1c2:1b7f:9fa0:ad74:311a:d89a:e0b1) |
| 2021-07-26 07:45:09 | → | merijn joins (~merijn@83-160-49-249.ip.xs4all.nl) |
| 2021-07-26 07:45:49 | × | euouae quits (~euouae@user/euouae) (Quit: Client closed) |
| 2021-07-26 07:46:35 | → | euouae joins (~euouae@user/euouae) |
| 2021-07-26 07:47:39 | → | burnsidesLlama joins (~burnsides@dhcp168-022.wadham.ox.ac.uk) |
| 2021-07-26 07:50:06 | × | azeem quits (~azeem@dynamic-adsl-94-34-48-122.clienti.tiscali.it) (Ping timeout: 258 seconds) |
| 2021-07-26 07:51:32 | → | azeem joins (~azeem@176.200.234.7) |
| 2021-07-26 07:52:33 | → | vysn joins (~vysn@user/vysn) |
| 2021-07-26 07:53:58 | × | azeem quits (~azeem@176.200.234.7) (Read error: Connection reset by peer) |
| 2021-07-26 07:54:12 | × | pottsy quits (~pottsy@2400:4050:b560:3700:81f2:32a:4b6a:ba03) (Quit: Leaving) |
| 2021-07-26 07:58:11 | → | azeem joins (~azeem@dynamic-adsl-94-34-48-122.clienti.tiscali.it) |
| 2021-07-26 08:03:18 | × | fossdd quits (~fossdd@sourcehut/user/fossdd) (Ping timeout: 265 seconds) |
| 2021-07-26 08:03:54 | → | fossdd joins (~fossdd@sourcehut/user/fossdd) |
| 2021-07-26 08:05:32 | → | hendursa1 joins (~weechat@user/hendursaga) |
| 2021-07-26 08:08:47 | × | hendursaga quits (~weechat@user/hendursaga) (Ping timeout: 244 seconds) |
| 2021-07-26 08:09:48 | × | euouae quits (~euouae@user/euouae) (Quit: euouae) |
| 2021-07-26 08:10:59 | → | kenran joins (~kenran@200116b82bf903005af8e44b0b810a86.dip.versatel-1u1.de) |
| 2021-07-26 08:11:06 | × | kenran quits (~kenran@200116b82bf903005af8e44b0b810a86.dip.versatel-1u1.de) (Client Quit) |
| 2021-07-26 08:17:47 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 2021-07-26 08:17:48 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 2021-07-26 08:17:51 | allbery_b | is now known as geekosaur |
All times are in UTC.