Ụlọ elu Kremlin dị na ogwe aka Hydra: ogbako na nkwekọ na nkesa Hydra 2020 kọmputa na Moscow

N'afọ gara aga na St. Petersburg e nwere mbụ Hydra ogbako, raara onwe ya nye na usoro ihe yiri ya na nkesa. Ndị mmeri mere ihe ngosi Dijkstra Prize и Turing ihe nrite (Leslie Lamport, Maurice Herlihy и Michael Scott), ndị okike nke mkpokọta na asụsụ mmemme (C ++, Go, Java, Kotlin), ndị mmepe nke ọdụ data ekesa (Cassandra, CosmosDB, Yandex Database), yana ndị okike na ndị nyocha nke algọridim na usoro data (CRDT, Paxos, chere -n'efu data Ọdịdị). N'ozuzu, n'oge a, ị nwere ike ịme ezumike, wedata windo IDE, mepee listi ọkpụkpọ na YouTube site na iji. akụkọ kacha mma Hydra 2019 - ka onye nhazi ọrụ chere ntakịrị.

N'ozuzu, ogbako dị otú ahụ emebeghị, ma ugbu a ọ ga-eme ọzọ. Ọzọ na akụkọ na Bekee, n'ihi na ọ dịghị asụsụ ka mma ikwu maka myirịta na-ekesa computing. Ọzọ n'oge ọkọchị, Julaị 10th na 11th, n'ihi na ndị ọkà okwu nwere oge nyocha na ịkụziri ihe, dịka ọmụmaatụ, na mahadum Cambridge, Rochester na St. Petersburg, na oge ndị ọzọ nke afọ abụghị maka ha.

Otú ọ dị, oge a ga-eme Hydra na Moscow, ebe ọtụtụ ndị sonyere ogbako bịara n'afọ gara aga ka ha gee ntị na akụkọ gbasara nkwekọrịta ekesa na ebe nchekwa azụmahịa. Hydra ọhụrụ ahụ nwere mmemme dị mgbagwoju anya karị, ndị na-ekwu okwu ọhụrụ yana ndị dike nke afọ gara aga, yana mmetụta nke maara nke ọma nke obi ụtọ nke hardcore yiri nke kesara n'etiti ndị sonyere na ụlọ nzukọ atọ.

Ka anyị tinye kaadị kaadị ozugbo na tebụl na ndị ọchịagha Byzantine chere ihu - anyị chọrọ ka mmemme Hydra ọhụrụ bụrụ nkọwa zuru oke ma dị iche iche. Oge ikpeazụ anyị ji mbọ aka kpụchaa, ugbu a, ka anyị gwuo ala na omimi. Nke a bụ isiokwu Hydra 2020 nwere ọdịiche gbasara afọ gara aga:

  Parallel systems:
* Algorithms & data structures
* Memory models
* Compilers, runtime
* Memory reclamation
* Testing & verification
* Hardware issues
* Non-volatile memory
* Transactional memory
* Scheduling algorithms & implementations
* Heterogeneous computing: CPU, GPU, FPGA, etc.
* Performance analysis, debugging, & optimization

  Distributed systems:
* Distributed computing
* Distributed machine learning/deep learning
* State machine replication & consensus
* Fault tolerance & resilience
* Testing & verification
* Hardware issues
* Blockchain & Byzantine fault tolerance
* Distributed databases, NewSQL
* Distributed stream processing
* Scheduling algorithms & implementations
* Cluster management systems
* Security
* Performance analysis, debugging, & optimization
* Peer-to-peer, gossip protocols
* Internet of things


Kedu otu esi ekwu maka ihe a niile na mmemme nke otu ọgbakọ? N'ezie, ọ dịghị mfe karịa ịnwale linearizability nke arụmọrụ na a na-egbuke egbuke ọhụrụ ekesa ụlọ ahịa na-eji Jepsen, ma anyị ga-agbalị.

Nke a bụ onye abanyelarị na mmemme:

Cindy Sridharan (Cindy Sridharan) - onye nrụpụta sistemụ kesara sitere na San Francisco, onye dere akwụkwọ dị mkpirikpi Nleba anya sistemu ekesa (were oyiri eletrọnịkị efu) na ewu ewu blọọgụ, ebe e nwere naanị otu isiokwu “Kachasị mma nke 2019 na Tech Talks"nwere ike ịzọpụta gị n'ime ụbọchị ole na ole, ma hapụ gị obi ụtọ. Na Hydra 2020, Cindy ga-agwa gị otu esi eme ya nwalee usoro ekesa, ọ bụrụgodị na ha na-echekwa steeti.


Michael Scott (Michael Scott) - nchọpụta si Mahadum nke Rochester, mara ndị mmepe Java niile dị ka onye okike algọridim na-anaghị egbochi ya na ahịrị na-emekọrịta ihe si na Java Standard Library. N'ezie, na Dijkstra Prize maka "Algorithms maka ngakọrịta nwere ike imekọrịta na ọtụtụ ihe nchekwa ebe nchekwa"na nwe ibe Wikipedia. N'afọ gara aga, Michael nyere akụkọ kacha mma (dị ka gị) na Hydra na nhazi data abuo, ma ugbu a ọ ga-ekwu maka ya oru ngo Hodor и ọrụ nchekwa na ebe nchekwa nkekọrịta, dị ka ndị yiri usoro.


Heidi Howard (Heidi Howard) - onye nyocha si Mahadum nke Cambridge, mara maka ịmepụta nkwekọrịta nkwekọrịta kesara algọridim Paxos na-agbanwe agbanwe, nakwa dị ka ọrụ na generalizing Flexible Paxos na Paxos ngwa ngwa. N'afọ gara aga, Heidi gwara otu o si arụ ọrụ na ọrụ Paxos ezinụlọ nke algọridim (otu n'ime akụkọ kacha mma), ma ugbu a, m ga-agbalị na-eje ije na mkpa ice n'etiti Ndị hụrụ Paxos na ndị na-akwado Raft - ma kesaa echiche ya nke algọridim dị mma.


Martin Kleppmann (Martin Kleppmann) bụ eleghị anya onye nchọpụta a ma ama na Mahadum Cambridge, na onye bụbu onye mmepụta nke nnukwu usoro data, onye dere akwụkwọ dị ịtụnanya nke doro anya na ya mere pụrụ iche na usoro kesara "Ịmepụta ngwa data na-akpa ike" N'afọ gara aga Martin kesara nsonaazụ ya nke nyocha CRDT ha, yana ihe anyị ga-agwa gị ugbu a anyị ga-ekwupụta ma emechaa.


Nikita Koval (Nikita Koval) bụ onye nrụpụta coroutine na otu Kotlin, onye nkuzi nkuzi na mmemme multi-threaded na ITMO, yana onye so na kọmitii mmemme nke ogbako Hydra (ee, nke isiokwu a gbasara). N'afọ gara aga Nikita kwuru banyere ịnwale multi-threaded data owuwu na JVM n'elu ikpo okwu na-eji Lin-Chekwa, na na Hydra 2020 ọ ga-agwa gbasara SegmentQueueSynchronizer - kwadoro iji Iris framework n'ihi na ilu Coq abstraction maka mmemme mmekọrịta primitives.


Soro ọkwa ọkwa anyị asynchronous: na mkpokọta a ga-enwe ihe dị ka akụkọ iri na abụọ na ọgbakọ ahụ, anyị ga-agwa gị maka ndị ọzọ n'oge na-adịghị anya. Ọzọkwa, n'ezie, na ogbako a ga-enwe mpaghara mkparịta ụka ebe ọ dị mkpa iji nwalee ndị ọkà okwu na ajụjụ na otu ma ọ bụ ọtụtụ eri ruo mgbe a ga-enwe nkwenye zuru ezu.

Ma ọ bụrụ na ị nwere chioma, Martin Kleppmann ga-abịanye aka n'akwụkwọ gị.

Ee, tupu ọgbakọ Hydra 2020, ya bụ Julaị 6-9, a ga-enwe SPTDC 2020 - ụlọ akwụkwọ okpomọkụ nke atọ na tiori na omume nke mgbakọ na-ekesa. Ọ ga-enye gị mmetụta nke siri ike ịnweta na ogbako, ya mere, anyị ga-ekwu banyere Ụlọ Akwụkwọ na post dị iche.

Gịnị ugbu a? Mbụ, soro akụkọ na Habré na netwọk mmekọrịta (Facebook, Вконтакте, Twitter).

Nke abuo, ọ bụrụ na ị nweelarị ọchịchọ a na-apụghị ịgbagha agbagha ịga ogbako ahụ, nyochaa weebụsaịtị, ị nwere ike ugbua ịzụta tiketi.

Nke atọ, echefula ohere gị na kọmitii mmemme ọgbakọ Hydra 2020 kparịta ụka na nkwupụta. Ndị otu PC ga-enwe obi ụtọ ịgwa gị gbasara isiokwu ọgbakọ n'ọdịnihu.

Hụrụ gị na Hydra!

isi: www.habr.com