Kubernetes don ML Workloads
Kubernetes tsarin buɗaɗɗen tushe ne wanda ke tsarawa ta atomatik, ma'auni, da kuma sake kunna shirye-shiryen kwantena a cikin tarin inji.
Dubawa
For machine learning, it lets teams pack GPU-hungry training jobs and latency-sensitive model servers onto shared hardware without babysitting individual servers.
Zurfafa nutsewa
An gina shi a asali a Google don gudanar da ayyukan gidan yanar gizo, Kubernetes yana ɗaukar gungun ku a matsayin babban tafki ɗaya na CPU, ƙwaƙwalwar ajiya, da GPUs, sannan ya yanke shawarar wacce injin ke tafiyar da kowane akwati. Ƙungiyoyin ML sun dogara da shi saboda ayyukan aiki sun fashe da tsada: guduwar horo na iya buƙatar GPUs takwas na sa'o'i shida, sannan ba komai. Kubernetes yana tsara jadawalin da ke kan kumburi tare da GPUs kyauta, kuma lokacin da aikin ya ƙare yana yantar da kayan aikin. Har ila yau, yana kiyaye sabar sabar da rai, tana sake kunna kwantena da suka faɗo da kuma yada kwafi a cikin injina don jurewa. Kayan aikin da aka gina a sama, kamar Kubeflow, Ray, da KServe, suna ƙara takamaiman nau'ikan ML irin su masu aikin horarwa da rarrabawa, daidaitawar hyperparameter, da ƙarshen ƙirar ƙirar autoscaling, don haka masana kimiyyar bayanai suna aiki tare da abstractions mafi girma maimakon raw YAML.
Fahimtar Fasaha
Kubernetes yana sanya GPUs ta hanyar plugins na na'ura waɗanda ke tallata albarkatu kamar nvidia.com/gpu, wanda mai tsarawa yayi daidai da buƙatun kwasfa. Taints da juriya suna kiyaye ayyukan CPU masu arha a kashe nodes na GPU masu tsada, yayin da masu zaɓen kumburi da ƙa'idodin alaƙa suna tura horo zuwa takamaiman kayan aiki. Don horar da GPU da yawa, masu aiki suna ƙirƙirar ƙungiyar kwasfan fayiloli waɗanda ke gano juna kuma suna gudanar da tsarin kamar PyTorch DDP ko Horovod, musayar gradients akan hanyar sadarwar tari ta amfani da NCCL.
Dabarun Tasiri
Kudin da kasafin kuɗi
Hukunce-hukuncen gine-gine suna haifar da aiki da tsadar aiki na shekaru.
Shawarwari masu haske
Ilimin fasaha yana taimaka wa ƙungiyoyi su zaɓi tari mai kyau, ba kawai sabon abu ba.
Kula da inganci
Zaɓuɓɓukan injiniya mafi kyau suna rage abin dogaro a cikin samarwa.
Makomar Kubernetes don ML Workloads
Yi tsammanin haɗin kai na ML mai ƙarfi: tsara tsarin ƙungiya wanda ke ƙaddamar da duk fastocin horarwa a lokaci ɗaya ko babu kwata-kwata, raba guntu da yanki na lokaci-lokaci don haka ayyukan haske da yawa suna raba kati ɗaya, da wuri mai sane da topology wanda ke mutunta haɗin haɗin gwiwar NVLink mai sauri. Ƙididdiga maras saɓani akan Kubernetes, madaidaicin maƙasudin ƙarshen zuwa sifili tsakanin buƙatun, yana girma. Kamar yadda balloon samfuri, masu tsara jadawalin suna ƙara daidaitawa a cikin gungu da gizagizai da yawa, da tsarin raba adalci na tushen layi kamar Kueue da Volcano sun zama ma'auni don sarrafa ƙarancin ƙarfin GPU.
Aiwatar da Gaskiyar Duniya
Gidan binciken bincike yana amfani da Ma'aikacin Horar da Kubeflow don ƙaddamar da 32-GPU PyTorch aikin horarwa da aka rarraba a cikin kuɗaɗe huɗu, sannan yana 'yantar da GPUs ta atomatik lokacin da ta haɗu.
Kamfanin kasuwancin e-commerce yana ba da samfurin shawarwarin sa tare da KServe, wanda ke ɗaukar kwafin kwafi yayin siyar da walƙiya da koma baya cikin dare.
Banki yana gudanar da ayyukan batch na dare kamar Kubernetes CronJobs, yana yi musu layi akan nodes na CPU don kada su yi gasa tare da zirga-zirgar rana.
Farawa yana amfani da Ray akan Kubernetes don gudanar da share fage na hyperparamita daidai gwargwado, yana jujjuya ɗimbin fakitin gwaji na ɗan gajeren lokaci akan abubuwan tabo don rage farashi.
Hatsari & Tsare-tsare
Haɓaka ma'auni ɗaya na iya ɓoye manyan raunin tsarin.
Sau da yawa ana raina kayan more rayuwa da kuma kuɗin kulawa.
Tsaro da gibin lura na iya girma yayin da tsarin ke ƙara haɓaka.
Taswirar Hanya
Ƙayyade latency, inganci, da maƙasudin farashi kafin aiwatarwa.
Alamar ma'auni a ƙarƙashin ainihin kaya da yanayin bayanai.
Kula da kayan aiki don kurakurai, ɗigo, da tasirin mai amfani.
Shirya bijirowa da hanyoyin mayar da martani kafin sikeli.
Ci gaba da Bincike
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Kubernetes for ML Workloads quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Jagora na gaba
Gwajin A/B don Samfuran ML
Tambayoyin da ake yawan yi
What is Kubernetes for ML Workloads?
Kubernetes tsarin buɗaɗɗen tushe ne wanda ke tsarawa ta atomatik, ma'auni, da kuma sake kunna shirye-shiryen kwantena a cikin tarin inji. Don koyon inji, yana barin ƙungiyoyi su tattara ayyukan horo na yunwar GPU da sabar ƙirar ƙira a kan kayan aikin da aka raba ba tare da kula da kowane sabar ba.
Menene aikin farko na mai tsara Kubernetes don ayyukan ML?
Mai tsara jadawalin ya dace da buƙatun albarkatun kwafsa (CPU, ƙwaƙwalwar ajiya, GPUs) akan nodes ɗin da ke akwai kuma yana sanya kwaf ɗin inda ya dace. Ba ya taɓa samfurin code ko bayanai.
Ta yaya Kubernetes yawanci ke ba da GPUs zuwa kwasfa?
Plugins na na'ura suna fallasa GPUs azaman albarkatun da za'a iya tsarawa (misali, nvidia.com/gpu), barin kwas ɗin su buƙace su da wadatar mai tsarawa.
Menene taints da haƙuri da aka fi amfani da su a cikin tarin ML?
Lalacewar tana korar kwas ɗin daga kumburi; kwasfa ne kawai tare da juriya mai dacewa zasu iya sauka a can. Wannan yana tanadi ƙarancin nodes na GPU don ayyukan da a zahiri ke buƙatar GPUs.
Wanne kayan aiki ne ke ƙara takamaiman damar ML kamar masu aikin horarwa a kan Kubernetes?
Kubeflow layers ML workflows akan Kubernetes, gami da masu gudanar da horo, bututun mai, da kuma daidaitawa, don haka ƙungiyoyi sun guje wa saitin gungu na ƙaramin matakin rubutu.
Me yasa Kubernetes ya dace da fashewar ML na ayyuka?
Horowa yana da kauri: GPUs da yawa a taƙaice, sannan babu. Kubernetes yana sanya aikin lokacin da albarkatun ke da kyauta kuma ya sake su bayan kammalawa, inganta amfani.