Ua paʻi ʻia ka paʻi 58th o ka pae o ka 500 o nā kamepiula hana kiʻekiʻe loa ma ka honua. I ka hoʻokuʻu hou, ʻaʻole i loli ka ʻumi kiʻekiʻe, akā 4 mau pūʻulu Lūkini hou i hoʻokomo ʻia i ka pae.
ʻO 19th, 36th a me 40th mau wahi i ka pae i lawe ʻia e nā hui Lūkini Chervonenkis, Galushkin a me Lyapunov, i hana ʻia e Yandex e hoʻoponopono i nā pilikia aʻo mīkini a hāʻawi i ka hana o 21.5, 16 a me 12.8 petaflops, kēlā me kēia. Holo nā pūʻulu i ka Ubuntu 16.04 a hoʻolako ʻia me nā kaʻina hana AMD EPYC 7xxx a me NVIDIA A100 GPUs: ʻo ka pūʻulu Chervonenkis he 199 nodes (193 tausani AMD EPYC 7702 64C 2GH cores a me 1592 NVIDIA A100 80G GPUs), Galushkin - EP136 tausani nodes AMD cores C 134 7702C 64GH a me 2 GPU NVIDIA A1088 100G), Lyapunov - 80 nodes (137 tausani cores AMD EPYC 130 7662C 64GHz a me 2 GPU NVIDIA A1096 100G).
Ma kahi o 43rd ka hui hou o Sberbank, ʻo Christofari Neo, e holo ana i NVIDIA DGX OS 5 (Ubuntu edition) a hōʻike i kahi hana o 11.9 petaflops. ʻOi aku ka nui o ka pūʻulu ma mua o 98 tausani mau cores computing e pili ana i ka AMD EPYC 7742 64C 2.25GHz CPU a hele mai me kahi NVIDIA A100 80GB GPU. Ua neʻe ka puʻupuʻu Sberbank Christofari i hoʻokō mua ʻia mai ka 61st a i ka 72nd wahi i ka pae i ka hapalua makahiki.
ʻElua mau puʻupuʻu home ʻē aʻe i noho i ka pae: Lomonosov 2 - neʻe mai 199 a 241 wahi (i ka makahiki 2015, ua lawe ʻia ka hui Lomonosov 2 i 31 wahi, a ʻo kona mua Lomonosov ma 2011 - 13 wahi) a me MTS GROM - neʻe mai 240 a 294 wahi . No laila, ua hoʻonui ʻia ka helu o nā hui kūloko i ka pae mai 3 a 7 ma mua o ʻeono mahina (no ka hoʻohālikelike ʻana, i ka makahiki 2020 aia nā ʻōnaehana home 2 i ka pae, ma 2017 - 5, a ma 2012 - 12).
No ka helu helu holoʻokoʻa, ʻo ka hui ʻo Fugaku Iapana, i kūkulu ʻia me ka hoʻohana ʻana i nā kaʻina hana ARM, e noho mau ana ma kahi mua. Aia ka pūʻulu Fugaku ma ka RIKEN Institute for Physical and Chemical Research a hāʻawi i kahi hana o 442 petaflops. Aia ka pūʻulu i nā nodes 158976 e pili ana i ka Fujitsu A64FX SoC, i hoʻolako ʻia me kahi 48-core Armv8.2-A SVE CPU (512 bit SIMD) me ke alapine o ka uaki o 2.2GHz. Ma ka huina, ʻoi aku ka nui o ka hui ma mua o 7.6 miliona mau cores processor (ʻekolu manawa ʻoi aku ma mua o ke alakaʻi ma mua), 5 PB o RAM a me 150 PB o ka waiho like ʻana e pili ana i ka Lustre FS. Hoʻohana ʻia ʻo Red Hat Enterprise Linux e like me ka ʻōnaehana hana. ʻO ka lōʻihi o nā uwea optical i hoʻohana ʻia e hoʻohui i nā nodes ma kahi o 850 mau kilomita.
ʻO ka lua o ka hui ʻo Summit cluster, i hoʻonohonoho ʻia e IBM ma ka Oak Ridge National Laboratory (USA). Holo ka puʻupuʻu iā Red Hat Enterprise Linux a loaʻa iā 2.4 miliona mau cores processor (22-core IBM Power9 22C 3.07GHz CPUs a me NVIDIA Tesla V100 accelerators i hoʻohana ʻia), e hāʻawi ana i ka hana o 148 petaflops, ʻaneʻane ʻekolu manawa i emi iho ma mua o ke alakaʻi ma. ka helu.
ʻO ke kolu o ka wahi i noho ʻia e ka hui ʻAmelika Sierra, i hoʻokomo ʻia ma ka Livermore National Laboratory e IBM ma ke kumu o kahi kahua e like me Summit a hōʻike i ka hana o 94 petaflops (ma kahi o 1.5 miliona mau cores). Hoʻohana ʻia ʻo Red Hat Enterprise Linux e like me ka ʻōnaehana hana.
Aia ma kahi ʻehā ka hui Kina Sunway TaihuLight, e hana ana ma ka National Supercomputer Center o Kina, me ka ʻoi aku o 10 miliona mau cores computing a hōʻike i kahi hana o 93 petaflops. Me nā hōʻailona hana like, hoʻopau ka hui Sierra i ka hapalua o ka ikehu e like me ka Sunway TaihuLight. ʻO ka ʻōnaehana hana ʻo ia ka hoʻoili Linux proprietary RaiseOS.
Ma ka ʻelima o ka hui ʻo Perlmutter cluster, i hana ʻia e HPE a aia ma ka National Energy Research Center ma USA. Aia ka pūʻulu he 761 tausani cores e pili ana i ka AMD EPYC 7763 64C 2.45GHz CPU a hāʻawi i kahi hana o 71 petaflops. ʻO ka ʻōnaehana hana ʻo Cray OS.
ʻO nā hiʻohiʻona hoihoi loa:
- Ka mahele ʻana ma ka helu o nā supercomputers ma nā ʻāina like ʻole:
- Kina: 173 (188 - ʻeono mahina aku nei). ʻO ka huina, hoʻopuka nā hui Kina i 17.5% o nā huahana āpau (ʻeono mahina i hala aku nei - 19.4%);
- USA: 149 (122). Kuhi ʻia ka nui o ka huahana ma 32.5% (ʻeono mahina i hala - 30.7%);
- Iapana: 32 (34);
- Kelemānia: 26 (23);
- Palani: 19 (16);
- Netherlands: 11 (16);
- UK: 11(11);
- Kanaka 11 (11);
- Rūsia 7 (3);
- Kōlea Hema 7 (5)
- Italia: 6 (6);
- Saudi Arabia 6 (6);
- Palakila 5 (6);
- Suedena 4 (3);
- Polani 4 (4);
- Australia, India, Switzerland, Finland: 3.
- I ka helu ʻana o nā ʻōnaehana hana i hoʻohana ʻia i nā supercomputers, ʻo Linux wale nō i koe no ʻehā a me ka hapa makahiki;
- Hāʻawi ʻia e ka hoʻohele Linux (ma nā brackets - ʻelua makahiki i hala aku nei):
- 51.6% (49.6%) ʻaʻole kikoʻī i ka hāʻawi ʻana,
- 18% (26.4%) hoʻohana i CentOS,
- 7.6% (4.8%) - RHEL,
- 7% (6.8%) - Cray Linux,
- 5.4% (2%) - Ubuntu;
- 4% (3%) - SUSE,
- 0.2% (0.4%) - Linux ʻepekema
- ʻO ka paepae hana haʻahaʻa loa no ke komo ʻana i ka Top500 i 6 mau mahina i hoʻonui ʻia mai 1511 a i 1649 teraflops (ʻekolu mau makahiki i hala aku nei, ʻo 272 mau puʻupuʻu wale nō i hōʻike i ka hana ʻoi aku ma mua o ka petaflop, ʻehā makahiki i hala - 138, ʻelima mau makahiki i hala - 94). No Top100, ua hoʻonui ka paepae komo mai 4124 a i 4788 teraflops;
- Ua hoʻonui ka nui o ka hana o nā ʻōnaehana āpau i ka helu mai 2.8 a i 3 exaflops i ka makahiki (ʻelua makahiki i hala aku nei he 1.650 exaflops, a ʻelima mau makahiki i hala - 566 petaflops). ʻO ka ʻōnaehana e pani ana i ka pae o kēia manawa aia ma kahi 433rd ma ka helu hope loa, a ma kahi 401st i ka makahiki ma mua;
- ʻO ka māhele nui o ka helu o nā supercomputers ma nā wahi like ʻole o ka honua penei: 226 supercomputers aia ma Asia (245 ʻeono mahina i hala), 160 ma ʻAmelika ʻĀkau (133) a me 105 ma ʻEulopa (113), 5 ma ʻAmelika Hema (6), 3 ma Oceania (2) a me 1 ma Africa (1);
- Ma keʻano he papa hana, aia nā Intel CPU i ke alakaʻi - 81.6% (ʻelua makahiki i hala aku nei he 94%), aia ʻo AMD ma kahi lua me 14.6% (0.6% !!), a aia ʻo IBM Power ma kahi ʻekolu me 1.4% ( he 2.8%). Aia ka ulu ikaika o nā pūʻulu e pili ana i nā kaʻina hana AMD; no ka laʻana, ua lako nā ʻōnaehana hou a pau i ka Top15 me nā AMD CPU.
- 26.6% (ʻelua makahiki i hala aku nei 35.6%) o nā mea hoʻohana a pau i loaʻa i 20 cores, 17.6% - 24 cores, 11.2% - 64 cores, 8.6% (13.8%) - 16 cores, 8.2% (11%) - 18 cores, 5.8 % (11.2%) - 12 kiko.
- 149 mai loko o 500 ʻōnaehana (ʻelua makahiki aku nei - 144) hoʻohana hou i nā accelerators a i ʻole coprocessors, me nā ʻōnaehana 143 e hoʻohana ana i nā pahu NVIDIA, 2 - Intel Xeon Phi (mai 5), 1 - PEZY (1), a me 1 AMD Vega GPU ;
- Ma waena o nā mea hana cluster, ua lawe ʻo Lenovo i kahi mua - 36.8% (ʻelua makahiki i hala aku nei 34.8%), ua lawe ʻo Inspur i ka lua o kahi - 11.6% (13.2%), ua lawe ʻo Hewlett-Packard Enterprise i ke kolu o kahi - 9% (7%), ukali ʻia e Sugon 7.8 % (14.2%), Atos - 7.2% (4.6%), Cray 6.4% (7%), Dell EMC 3.2% (2.2%), Fujitsu 3% (2.6%), NVIDIA 2.4 (1.2%), NEC 2% , Huawei 1.4% (2%), IBM 1.4% (2.6%), Penguin Computing - 1.4% (2.2%). ʻEhiku mau makahiki i hala aku nei, ʻo ka hāʻawi ʻana i waena o nā mea hana penei: Hewlett-Packard 36%, IBM 35%, Cray 10.2% a me SGI 3.8%;
- Hoʻohana ʻia ʻo Ethernet e hoʻohui i nā nodes ma 49.4% (ʻelua makahiki i hala aku nei 52%) o nā pūʻulu, hoʻohana ʻia ʻo InfiniBand i 33.6% (28%) o nā pūʻulu, Omnipath - 8.4% (10%). Ke nānā nei i ka hana holoʻokoʻa, ʻo nā ʻōnaehana hoʻokumu InfiniBand no 43.3% o ka hana holoʻokoʻa o Top500, ʻoiai ʻo Ethernet no 21.3%.
I ka wā e hiki mai ana, manaʻo ʻia e paʻi ʻia kahi hoʻokuʻu hou o ka Graph 500 o nā ʻōnaehana cluster, e nānā ana i ka loiloi ʻana i ka hana o nā kahua supercomputer e pili ana i ka hoʻohālikelike ʻana i nā kaʻina hana kino a me nā hana no ka hoʻoili ʻana i ka nui o nā ʻikepili i loaʻa i loko o ia mau ʻōnaehana. Hoʻohui ʻia ka Green500, HPCG (High-Performance Conjugate Gradient) a me HPL-AI me ka Top500 a hōʻike ʻia i ka pae nui Top500.
Source: opennet.ru