I-inthanethi Yezinto Ecosystem ikhula ingahloliwe, futhi isofthiwe yomthombo ovulekile idlala indima ehamba phambili. Kusukela kumapulatifomu e-IoT kuya kumathuluzi wedatha namamodeli we-AI onqenqemeniKunohlu lwezinketho ezikuvumela ukuthi wakhe izixazululo ezinamandla ezinezindleko ezizwakalayo kanye nokuvumelana nezimo okukhulu. Uma ucabangela ukuthi yibuphi ubuchwepheshe obuzobusebenzisa, nali umhlahlandlela ophelele nosebenzayo onolwazi olufanele kakhulu.
Ngaphambi kokuthi singene esihlokweni, kubalulekile ukucacisa imiqondo ethile. “Umthombo ovulekile” awufani ncamashi “nesoftware yamahhala”Futhi ukukhetha phakathi kohlaka lomthombo ovulekile noma inkundla yangokwezifiso kungenza umehluko ekuvikelekeni, izindleko, nokuvela kwephrojekthi yakho. Ngaphezu kwalokho, uma usebenza ngamadivayisi anezinsizakusebenza, uzoba nentshisekelo yokufunda ukuthi yimaphi amamodeli we-AI angasindi enza kahle emaphethelweni.
Isofthiwe yamahhala iqhathaniswa nesofthiwe yomthombo ovulekile
Amagama avame ukudideka, kodwa anama-nuances abalulekile. Isofthiwe yamahhala ibeka phambili inkululeko yomsebenzisiInkululeko yokusebenzisa, ukufunda, ukulungisa, nokusabalalisa kabusha uhlelo, ngenkululeko emine ye-FSF njengokugxila kwayo okuyinhloko. Umthombo ovulekile, ngakolunye uhlangothi, ugcizelela ukutholakala kwekhodi yomthombo ukuze kubukwe, kulungiswe, futhi kusatshalaliswe. nakuba amalayisensi angase abeke imibandela mayelana nokusabalalisa kabusha noma izinguquko futhi awufuni njalo imigomo efanayo nokunyakaza kwesofthiwe yamahhala.
Ngamanye amazwi, Ukugxila kwesoftware yamahhala kumalungelo abasebenzisiNakuba umthombo ovulekile ugxile kukhodi evulekile nokusebenzisana, kukhona ukugqagqana, kunjalo, kodwa amafilosofi nezinjongo zabo akufani.
Iyini ngempela inkundla ye-IoT?
Inkundla ye-IoT ingumgogodla oxhumanisa amadivaysi, idatha, kanye nezinhlelo zokusebenza. Ihlinzeka ngengqalasizinda namabhulokhi wokwakha ukuthuthukisa isevisi yakho: ukuxhumana namadivayisiUkuphatha nokuhlinzeka, ukuphepha, ukuhlanganiswa kwamafu, namathuluzi okuhlaziya, phakathi kokunye. Empeleni, Ikuvumela ukuthi ugxile ekusebenzeni kwebhizinisi ngaphandle kokuvuselela isondo ekuxhumekeni nasekusebenzeni.
Amamodeli we-AI angasindi onqenqemeni lwe-IoT
Ekusetshenzisweni onqenqemeni ngenkumbulo elungiselelwe, ibhethri, ne-CPU, Amamodeli aqinile futhi asebenzayo enza umehlukoEzinye ezigqamayo ngebhalansi yazo phakathi kosayizi namandla yilawa: I-Meta Llama 3.1 8B (ukusebenza kahle kwezilimi eziningi), i-GLM-4-9B-0414 (izingcingo ezisebenzayo nokuhlanganiswa namathuluzi) kanye ne-Qwen2.5-VL-7B (ubuhlakani be-multimodal obunombono). Lezi zindlela ziyisihluthulelo sokuvumela izinqumo zendawo ngaphandle kokuncika njalo ngefu.
Amapulatifomu e-IoT yomthombo ovulekile nezihloko ezihlobene
I-DeviceHive
I-DeviceHive inikezela ngohlaka lwe-M2M lokuxhuma amadivayisi nokuphatha izinsiza zamafu ngokugxila ku-Big Data. Ihlanganisa iphaneli yewebhu elula kakhulu. ukudala amanethiwekhi, ukusebenzisa imithetho yezokuphepha nokuqapha, kanye namaphrojekthi ayisampula kanye nesibonelo sokuhlola ku-inthanethi. Isiphakamiso sabo sihlanganisa ibhuloho phakathi kwamafu, amasistimu ashumekiwe nezinhlelo zokusebenza zeselula ngendlela efanele elinganiselayo.
ThingSpeak
I-ThingSpeak, exhunywe eduze ne-MathWorks, igqama ngezibalo zayo zisebenzisa i-MATLAB kudatha yenzwa eseduze yesikhathi sangempela. Phakathi kwemisebenzi yayo Lokhu kufaka phakathi ukuhlanganisa nokuhlaziywa kokusakaza bukhoma, ukurekhodwa kweziteshi zomphakathi ukuze zisetshenziswe ngasese, ukwabelana kwesiteshi, ukubonwa, ukubuyekezwa nge-REST API ne-MQTT, amathuluzi aku-inthanethi e-MATLAB, kanye izexwayiso nge-TimeControl kuqalwe izehlakalo.
I-Mainflux
I-Mainflux iyinqwaba ye-Golang ekhanya ngokuxhuma nokuphatha kwayo. Isekela i-HTTP, i-MQTT, iWebSocket ne-CoAPIhlanganisa ukunikezwa nokuphathwa kwedivayisi, ukuthunyelwa kwe-Docker, kanye ne-Kubernetes orchestration. ukuphepha ngokhiye be-API enziwa ngendlela oyifisayo kanye ne-JWT ngokufinyelelafuthi kusiza ukunciphisa i-OPEX. Ingafakwa emagcekeni, endaweni eyingxubevange, noma emafini.
Thinger.io
I-Thinger.io iyinkundla yamafu esetshenziswa ngeziqukathi ze-Docker. Ihlanganisa izinhlobonhlobo zehadiwe. (Arduino IDE, Linux, Sigfox, ARM Mbed boards), inikeza ikhonsoli yokuphatha esebenziseka kalula, esakaza kuma-websockets, amadeshibhodi edatha yesikhathi sangempela, Izinhlelo zokusebenza zeselula ze-iOS ne-Android kanye nokuzenzakalelayo kwe-IFTTT kumadivayisi amaningi. Nakuba ukubonakala kwayo kungekona okukhanga kakhulu, Indlela yabo "yohlelo olusebenzayo" ingamandla abo amakhulu..
I-Google Cloud IoT
Yize ingewona umthombo ovulekile kanjalo, inkundla yeGoogle inikeza izingcezu eziwusizo zezakhiwo ezixhunyiwe ze-IoT. Sebenzisa ibhuloho le-MQTT Ngokuxhumeka, ixhuma kusilinganisi senethiwekhi sommeleli wangaphandle. Izinhlelo zokusebenza ze-IoT zingasebenzisa idatha nge-Pub/Sub noma zisebenzise isixhumi se-Dataflow's MQTT, kuyilapho uhlelo lunikeza izinsiza zokuphatha ezibalulekile: imininingwane, ukuqinisekiswa nokugunyazwaUkucushwa kwedivayisi nokubhaliswa, injini yemithetho, nokuphathwa kokubuyekeza nakho kuyizici ezivamile. Amawele edijithali, izexwayiso nezibalo ku-ecosystem yayo.
I-OpenRemote
I-OpenRemote isiza ukuhlanganiswa kwezihlanganisi, abakhiqizi, kanye nokuphatha ngezivumelwano ezifana ne-HTTP, SNMP, MQTT, noma i-Bluetooth. Amandla akho Ihlanganisa ukubukwa kwemephu nempahla, imodeli yempahla enezinhlobo ezichazwe umsebenzisi, izivumelwano zama-ejenti alungisekayo, isixhumi esibonakalayo esinesithombe semithetho ye-IF-THEN, amadeshibhodi, ukuxhumana phakathi kwezimo ze-Edge kanye nendawo emaphakathi kanye nokuphathwa kwabaqashi abaningi. Isebenza ku amabuthoamandla, ukuphathwa okuhlangene, izakhiwo, amadolobha, izikhumulo zezindiza noma ukunakekelwa kwezempilo.
Umthombo ovulekile noma inkundla yangokwezifiso?
Nakuba amapulatifomu amaningi avulekile angase abonakale afana ephepheni, ukukhetha kuncike ezinhlosweni, isabelomali, ubungozi, kanye nethimba. Umthombo ovulekile wonga isikhathi futhi unikeza ukuguquguqukaNokho, ayinazo izinselele zayo: iminikelo yekhwalithi engalingani, ukunakekelwa okuntekenteke nokuvikeleka, kanye nobumfihlo bedatha obubaluleke kakhulu. izindleko azihlali ziphansi kanye nesethi yezici ezijwayelekile okungenzeka zingahambisani kahle nemigomo yakho. Ngaphezu kwalokho, Ngokuvamile akuzona ezabasebenzisi abangenalwazi..
Uma ukhetha inkundla yangokwezifiso, impikiswano enkulu ingukulawula. Yiba nekhono eliphelele lokuklama, ukuphepha, nokusebenza Ivumela ukuhlanganiswa namasistimu akhona, ivala ukuvikelwa kokuphela, futhi ithuthukise ukusebenza, ukwethembeka, kanye nokuqina. Le ndlela, lapho isetshenziswa kahle, kwandisa ukubuya futhi ulungiselele isisombululo sakho ukuze sikhule endaweni eshintsha ngokushesha.
Isixazululo se-eManager kanye ne-suite yamathuluzi
I-eManager idalwe njengesilawuli se-IoT sezimboni sabachwepheshe besoftware abadingayo izinto ezihlukahlukene, i-multi-protocol hardware lapho kuzosetshenziswa kalula amaphrojekthi. Indawo yabo yokugcina ihlanganisa uhlu oluyi-Top 8 lokutholwa, ukubonwa, kanye nesoftware yokugcina, eyenzelwe faka amacala omhlaba wangempela ekukhiqizeni ngokushesha
Ukuthola nokubona ngeso lengqondo
I-Chirpstack inikeza isitaki somthombo ovulekile esiphelele samanethiwekhi e-LoRaWAN ngokuhumusha komlayezo, ukuhlanganiswa kwamafu, ukuphathwa kwezempilo, i-inventory, ukwenza kusebenze izinzwa nokuthunyelwa kwedatha kumadivayisiIhlinzeka ngesixhumi esibonakalayo sewebhu sabasebenzisi, izinhlangano, izinhlelo zokusebenza, namadivayisi, futhi iveza i-gRPC ne-REST API. Idatha ingageleza nge-MQTT/HTTP futhi ibhalwe ku-InfluxDB.
I-Node-RED, ngakolunye uhlangothi, ingummese Wezempi WaseSwitzerland we-IoT wokuhlela ukugeleza ngekhodi encane. Uhlelo lwayo olubukwayo lukwenza kube lula ukuxhuma ihadiwe, ama-API, namasevisi futhi uziguqule ngesikhathi sangempela, ukwenze kube ukukhetha okuphezulu ku-Industry 4.0. Uma ufuna ukushesha ngaphandle kwejika lokufunda eliwumqansa, Kufanele ube nakho..
Isitaki se-TICK sihlanganisa i-Telegraf (i-ejenti yamamethrikhi nemicimbi), i-Kapacitor (ukucubungula kwesikhathi sangempela) ne-Chronograf (i-InfluxData interface), ne-InfluxDB njengesisekelo sayo. Ndawonye bakha isixazululo sokugcina ukuthwebula, ukuqapha, ukubona ngeso lengqondo nokwenza okuzenzakalelayo kuchungechunge lwesikhathi ngokubhala okusindayo nomthwalo wemibuzo.
U-Grafana uqedela umbuthano wokubuka ngamadeshibhodi anothile nezexwayiso. Ilungele ukusebenzisa idatha ye-InfluxDB. neminye imithombo, ukwakha amadeshibhodi akhangayo ngaphandle kwezinkinga. Uma udinga ukuthola amaphethini noma okudidayo nqunuNangu umngane wakho.
Yolwazi
I-InfluxDB, ngaphakathi kwesitaki se-TICK, ithuthukiselwe uchungechunge lwesikhathi: Kubhalwe okuningi, imibuzo esheshayoUkuqapha, amamethrikhi, izinzwa, nezibalo zesikhathi sangempela. Uma umzuzwana ngamunye ubalulekile, ukwakheka kwayo kwenza umehluko.
UMariaDB uthola okuhle kakhulu kwe-MySQL futhi wengeza ukuthuthukiswa: ukulondoloza isikhashana semibuzo eyinkimbinkimbiUkuphathwa kokuxhumeka okusha, ukufinyelela kweqoqo, nokusekelwa kwezigaba ezithuthukile nezakhiwo. Uma uphuma ku-MySQL ecosystem, Uzoyithola ijwayelekile futhi inamandla.
I-PostgreSQL iwuphawu lokuma kusizindalwazi esihlobene somthombo ovulekile, esinokuvumelana okuphezulu, izinhlobo zedatha ezihlukahlukene, umumo wezinto kanye nokuhambisana kwenkundla. Isekela i-SQL yedatha yobudlelwane kanye ne-JSON yedatha engahlobene., okwenza kube nezimo kakhulu kumaphrojekthi ayingxube.
I-SQLite ilungele amasistimu ashumekiwe kanye nehadiwe enesizotha: engasindi, iyasebenza futhi iyashesha ukuze uthole isitoreji sasendaweni esingelula. Uma idivayisi iwukukhathazeka okuyinhloko futhi izinsiza zikhawulelwe, Ikukhipha ezindaweni eziminyene ezingaphezu kweyodwa..
Yini entsha ku-suite
Ububanzi be-eManager bubuyekeze isofthiwe yabo yaba yi-Node-RED 2.1.4 kanye ne-Yocto Dunfell, ehlanganisa nephrothokholi ye-UPnP nokunye ukuthuthukiswa. Uzuza ekusebenzeni, ubuyekeze ukuphepha, ukulungisa amaphutha, nokungenaukucwenga umuzwa wokufuna ukuthunyelwa kwezimboni. Uma ufuna imininingwane emihle, bheka imibhalo esemthethweni futhi uhlale unolwazi lwakamuva nge-newsletter yabo.
Umthombo ovulekile: ukuthi uhlanganisani nokuthi kungani ubalulekile
Sibiza umthombo ovulekile noma yiliphi ithuluzi okushicilelwe ikhodi yalo yomthombo futhi ingabukwa, ilungiswe futhi isatshalaliswe kabushaLokhu kunika amandla ukuthuthukiswa okusebenzisanayo nokusobala lapho umphakathi unikela khona ngokulungiswa kweziphazamisi, izici ezintsha, nokuthuthukiswa kwekhwalithi. Amalayisense ngokuvamile avumela lokhu kusetshenziswa. ukukhuthaza ukusungulwa kanye nokuzivumelanisa nezimo ezidingweni ezithile.
Kunezibonelo eziningi: i-Linux kumasistimu wokusebenza, iMozilla Firefox ezipheqululini, noma i-LibreOffice kuma-office suite. Iphethini iyaziphindaUma kuphakama inkinga enkulu futhi singekho isixazululo, iphrojekthi evulekile ibonakala izobhekana nayo.
Izinsiza zomthombo ovulekile ezifakiwe zedatha, i-backend, nokuningi
1) Apache Hudi
I-Hudi inikeza uhlaka lwesitoreji esikhuphukayo ngesikhathi sangempela kanye nokucubungula ngaphezulu kwe-Hadoop ne-Spark. Ukugxila kwabo ekucasuleni, ekususeni, nasekukhuphukeni Lokhu kuyenza ifanelekele ukungeniswa kwedatha okuqhubekayo nokuhlaziya okusebenzisanayo; ihambisana namachibi edatha namachibi, okuvumela imibuzo ye-latency ephansi ngamavolumu amakhulu. Ihlanganisa ne-Spark, Flink, Presto, StarRocks, noma i-Amazon Athena.
2) I-Apache Iceberg
I-Iceberg inikeza ifomethi yethebula lokuthengiselana ngokubhala kwe-athomu, izifinyezo, ukufundwa okuhle nokuhlukanisa/ukuhlungaIsebenza ne-Spark, Hive, Presto nezinye izinjini (ClickHouse, Dremio, StarRocks), ukuxazulula Amabhodlela e-HDFS/Hive kumadathasethi amakhulu kanye nokwenza lula ukuvela kwesikimu, ukuminyana kanye nokuhlehla emuva.
3) I-Apache Superset
I-Superset iyinkundla yezibalo zokuzisiza amaqembu amaningi adinga ukuyikala. Xhuma nge-SQL, izindawo zokugcina idatha namachibi edathaInikeza umakhi weshadi kanye ne-SQL IDE, futhi isekela yonke into kusukela kubha kanye namashadi ophayi kuya ekubukweni kwe-geospatial okuthuthukisiwe. Uma iphrojekthi yakho inengxenye ye-BI eqinile, Kukongela isikhathi namakhanda..
4) Bun
I-Bun igxile ethuluzini lesikhathi sokusebenza esisodwa, umphathi wephakheji kanye nephakheji ye-JS yohlangothi lweseva. Amandla ayo amakhulu ukusebenza (sibonga ngokwengxenye ku-Zig) kanye nolwazi oluhlanganisiwe, okwenza kube enye indlela engathi sína ye-Node ne-Deno. Uma ungathanda ukushefa imizuzwana ku-CLI nezikhathi zokwakha, Inikeze i-spin.
5) Claude 2
Lo msizi we-Anthropic uphatha izimo ezinkulu (kufika kumathokheni angu-100.000) futhi abhale kahle noma aguqule umbhalo ngezilimi eziningi. Ungafingqa, ukhiphe, ubhale kabusha, futhi uphendule ngokusekelwe kokuqukethwe, futhi uqonda izilimi zokuhlela ezijwayelekile. Uqeqeshwe ngaphansi komgomo we-HHH (Uwusizo, Uthembekile, Awunangozi). ayithambekele kakhulu ekuphumeni okuyingozi futhi ayiziqeqeshi ngedatha yakho noma ixhumane ne-inthanethi ukuze iphendule.
6) I-CockroachDB
Isatshalaliswa, ithobela i-ACID, futhi itholakala kakhulu isizindalwazi se-SQL esiphindaphindwayo futhi ukukala okuvundlile kokufunda nokubhalaIlungele ukusetshenziswa okuphezulu noma izinhlelo zokusebenza zokuphakelwa kwezifunda eziningi, kunciphisa ukubambezeleka futhi kusiza ngokuthobela imithetho. Iwusizo ezinhlanganweni ezifana ne-Netflix nezikhungo zezezimali. Bayisebenzisa ekukhiqizeni okukhulu.
7) I-CPython (intuthuko yakamuva)
I-Python 3.11 kanye ne-3.12 ilethe i-a ukweqa okubambekayo ekusebenzeni komdlali oyinkombakuzuzisa amaphrojekthi amaningi ngaphandle kokushintshwa kwekhodi. Emkhathizwe, izinhlelo zokukhuluma ne-GIL zivula umnyango ukufana kweqiniso ku-multiprocessing kanye nemali engenayo eyengeziwe.
8) I-DuckDB
Injini yokuhlaziya eshumekiwe nge ukubulawa kwekholomu, ukufana nokusetshenziswa okuphansiIlungele imibuzo eyinkimbinkimbi kanye nokuhlola okusebenzisanayo kumakhompyutha aphathekayo noma amasistimu ashumekiwe. Ihambisana ne-SQL ejwayelekile, okwenziwayo kwe-ACID, nokuhlanganiswa nama-panda ne-dplyr. Yehlisa ukungqubuzana phakathi kokuhlaziya nokusebenzisa.
9) I-HTML ne-Hyperscript
Baphakamisa ukuthi kuhambe "i-throttle egcwele" nge-HTML evamile: I-HTMX ithatha indawo ye-JavaScript ephindaphindayo Izibaluli ezimemezelayo ze-AJAX, izifundazwe, nedathaI-HyperScript yenza lula ukuvumelanisa kanye ne-DOM nge-syntax yesitayela se-HyperCard. Ndawonye bayanikela okunye okusangulukile kunezinhlaka ezisebenzayo uma ufuna isivinini ngaphandle kokugcwala.
10) I-Istio
I-mesh yesevisi yokugcina: isiza umzila nokulinganisa ukulayisha, ukubonwa okuningiliziweUkubethela, ukufakazela ubuqiniso, nokugunyazwa phakathi kwama-microservices. Ihlanganisa ne-Kubernetes futhi ihlukanise inethiwekhi nokuphepha yekhodi, izinqubomgomo ezilinganisa ekusetshenzisweni okuyinkimbinkimbi.
Umphakathi kanye nemikhuba emihle
Uma usebenza nge-AI kanye namathuluzi wokuzenzakalela, kuwumqondo omuhle ukubamba iqhaza emiphakathini eyabelana ngekhodi, amathiphu, nezinqubo ezihamba phambili. Kukhona ama-subreddits agxile ekuhlelweni nge-ChatGPT lapho ukusebenzelana, amathiphu okusetshenziswa, namaphrojekthi aphelele ashicilelwe; ungakhohlwa ukufunda imithetho ngaphambi kokuthumela ukuze ulondoloze indawo ewusizo nehloniphekile.
Izinsiza nokwanda
Uma ungathanda ukufunda okwengeziwe ngamathuluzi omthombo ovulekile, kunohlu olubanzi olunamaphrojekthi amaningi ahlanganisa ukuthuthukiswa, idatha, izibalo, i-AI, ne-ML. Ukukhethwa kwangaphambilini kusukela koku-1 kuye ku-10 futhi iqhubekela phambili ezihlokweni ezilandelayo. Ukuze uthole ulwazi oluthe xaxa, ungadawuniloda amadokhumenti neziqondiso zobuchwepheshe, njengemibiko yasenyuvesi namaphepha amhlophe, Banwebeka emicabangweni yamachibi edatha, izindawo zokugcina izimpahla, nezindawo zamachibi.Nasi isibonelo sereferensi ngefomethi ye-PDF. etholakalayo ukuze kuboniswaneFuthi-ke, Silandele ku-Twitter naku-LinkedIn Uma ufuna ukuhlala usesikhathini nezindaba.
Uma ubheka yonke into, kubonakala kuyindwangu yezobuchwepheshe eqinile kakhulu: vula amapulatifomu e-IoT anezingxenye ezivuthiwe (I-DeviceHive, ThingSpeak, Mainflux, Thinger.io, OpenRemote), izinketho zamafu ezilingana kahle nokugeleza komsebenzi okuvulekile (i-Google Cloud IoT), i-industrial suite efana ne-eManager enamathuluzi abalulekile (Chirpstack, Node-RED, TICK, Grafana, InfluxDB, MariaDB, PostgreSQL, SQLite, amaphrojekthi we-Idi, umthombo we-backset), kanye nesitshalo sedatha evulekile I-Bun, Claude 2, CockroachDB, CPython, DuckDB, HTML/Hyperscript, Istio) ukuthi Bahlanganisa yonke into kusukela ekungeneni kuye ekubukweni nasekubuseniNgakho konke lokhu, unezinto eziningi ezanele zokuklama isixazululo se-IoT esiqinile, esivikelekile, esikalayo esilungele ukukhula ngaphandle kokulahlekelwa ukulawula.
