Node.js’te \u00c7oklu \u0130\u015f Par\u00e7ac\u0131\u011f\u0131 Kullan\u0131m\u0131: Kapsaml\u0131 Rehber<\/title><br \/>\n<body><\/p>\n<p>Node.js’in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js’te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!<\/p>\n<h2>Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?<\/h2>\n<p>Node.js, geli\u015ftiriciler aras\u0131nda h\u0131zla pop\u00fclerlik kazanm\u0131\u015f bir JavaScript \u00e7al\u0131\u015fma zaman\u0131 ortam\u0131d\u0131r. Ba\u015flang\u0131\u00e7ta, Node.js’in en dikkat \u00e7ekici \u00f6zelliklerinden biri, tek i\u015f par\u00e7ac\u0131kl\u0131 (single-threaded) yap\u0131s\u0131yd\u0131. Bu mimari, genellikle sunucu taraf\u0131 uygulamalar i\u00e7in bir avantaj olarak g\u00f6r\u00fcl\u00fcyordu \u00e7\u00fcnk\u00fc karma\u015f\u0131k senkronizasyon mekanizmalar\u0131na olan ihtiyac\u0131 azalt\u0131yor ve geli\u015ftirme s\u00fcrecini basitle\u015ftiriyordu. Ancak, bu tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131, \u00f6zellikle CPU yo\u011fun i\u015flemler s\u00f6z konusu oldu\u011funda baz\u0131 \u00f6nemli s\u0131n\u0131rlamalar\u0131 da beraberinde getirmektedir. Node.js, bu tek i\u015f par\u00e7ac\u0131kl\u0131 modelle birlikte, non-blocking I\/O (giri\u015f\/\u00e7\u0131k\u0131\u015f) ve bir Event Loop (Olay D\u00f6ng\u00fcs\u00fc) mekanizmas\u0131 kullanarak y\u00fcksek e\u015f zamanl\u0131l\u0131k (concurrency) sa\u011flamay\u0131 hedefler. Peki, bu tam olarak ne anlama geliyor ve uygulamalar\u0131m\u0131z\u0131 nas\u0131l etkiliyor?<\/p>\n<p>Temel olarak, Node.js’deki tek i\u015f par\u00e7ac\u0131\u011f\u0131, t\u00fcm JavaScript kodunuzun tek bir ana i\u015f par\u00e7ac\u0131\u011f\u0131nda y\u00fcr\u00fct\u00fcld\u00fc\u011f\u00fc anlam\u0131na gelir. Bu i\u015f par\u00e7ac\u0131\u011f\u0131, Event Loop ad\u0131 verilen \u00f6zel bir mekanizma taraf\u0131ndan y\u00f6netilir. Event Loop, gelen istekleri dinler, I\/O i\u015flemlerini (dosya okuma\/yazma, a\u011f istekleri, veritaban\u0131 sorgular\u0131 vb.) i\u015fletim sistemine delege eder ve bu i\u015flemler tamamland\u0131\u011f\u0131nda geri d\u00f6n\u00fc\u015fleri i\u015fler. \u00d6nemli olan nokta \u015fudur: I\/O i\u015flemleri genellikle zaman al\u0131c\u0131d\u0131r, ancak Node.js bu i\u015flemleri bloklamadan (non-blocking) yapar. Yani, bir veritaban\u0131 sorgusu g\u00f6nderdi\u011finizde, ana i\u015f par\u00e7ac\u0131\u011f\u0131 bu sorgunun tamamlanmas\u0131n\u0131 beklemez; bunun yerine, Event Loop di\u011fer g\u00f6revleri i\u015flemeye devam eder ve sorgu tamamland\u0131\u011f\u0131nda bir geri \u00e7a\u011f\u0131rma (callback) fonksiyonu \u00e7al\u0131\u015ft\u0131r\u0131l\u0131r. Bu yakla\u015f\u0131m, Node.js’in binlerce e\u015f zamanl\u0131 ba\u011flant\u0131y\u0131 verimli bir \u015fekilde y\u00f6netmesini sa\u011flar.<\/p>\n<p>Ancak, bu modelin bir zay\u0131f noktas\u0131 vard\u0131r: CPU-bound (i\u015flemciye ba\u011fl\u0131) i\u015flemler. CPU-bound i\u015flemler, yo\u011fun hesaplama gerektiren ve i\u015flemciyi uzun s\u00fcre me\u015fgul eden g\u00f6revlerdir. \u00d6rne\u011fin, karma\u015f\u0131k bir matematiksel hesaplama, b\u00fcy\u00fck bir dosyan\u0131n \u015fifrelenmesi veya video d\u00f6n\u00fc\u015ft\u00fcrme gibi i\u015flemler CPU-bound kategorisine girer. E\u011fer ana i\u015f par\u00e7ac\u0131\u011f\u0131nda b\u00f6yle bir CPU-bound i\u015flem \u00e7al\u0131\u015ft\u0131r\u0131rsan\u0131z, Event Loop kilitlenir (blocked) ve di\u011fer t\u00fcm gelen istekler veya bekleyen I\/O geri d\u00f6n\u00fc\u015fleri bu i\u015flem bitene kadar i\u015flenemez. Bu durum, uygulaman\u0131z\u0131n donmas\u0131na, gecikmelere ve genel performans d\u00fc\u015f\u00fc\u015f\u00fcne yol a\u00e7ar. Kullan\u0131c\u0131lar, uygulaman\u0131n yan\u0131t vermedi\u011fini veya yava\u015flad\u0131\u011f\u0131n\u0131 fark edebilirler. \u0130\u015fte tam da bu noktada, Node.js’in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak i\u00e7in \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kavram\u0131 devreye girer. Bu sayede, yo\u011fun hesaplamal\u0131 g\u00f6revler ana i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 me\u015fgul etmeden, ayr\u0131 i\u015f par\u00e7ac\u0131klar\u0131nda paralel olarak y\u00fcr\u00fct\u00fclebilir hale gelir.<\/p>\n<h2>Node.js’te \u00c7oklu \u0130\u015f Par\u00e7ac\u0131\u011f\u0131na Neden \u0130htiya\u00e7 Duyulur? Ger\u00e7ek D\u00fcnya Senaryolar\u0131 Nelerdir?<\/h2>\n<p>Node.js’in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131, I\/O yo\u011fun uygulamalar i\u00e7in m\u00fckemmel bir se\u00e7imdir. Ancak, modern web uygulamalar\u0131 ve servisleri giderek daha karma\u015f\u0131k hale geliyor ve sadece I\/O i\u015flemleriyle s\u0131n\u0131rl\u0131 kalm\u0131yor. \u00d6zellikle b\u00fcy\u00fck \u00f6l\u00e7ekli veri i\u015fleme, yapay zeka modellerinin \u00e7al\u0131\u015ft\u0131r\u0131lmas\u0131, medya d\u00f6n\u00fc\u015ft\u00fcrme veya karma\u015f\u0131k raporlama gibi g\u00f6revler, ciddi miktarda CPU g\u00fcc\u00fc gerektirebilir. Bu t\u00fcr CPU-bound i\u015flemler, Node.js’in ana i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 kilitleyerek t\u00fcm uygulaman\u0131n performans\u0131n\u0131 olumsuz etkileyebilir. Bu nedenle, Node.js ekosistemi, bu t\u00fcr senaryolarla ba\u015fa \u00e7\u0131kabilmek i\u00e7in \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 yeteneklerini sunmu\u015ftur. Peki, ger\u00e7ek d\u00fcnyada hangi durumlarda \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131na ihtiya\u00e7 duyar\u0131z ve bu, uygulamalar\u0131m\u0131z\u0131n mimarisine nas\u0131l bir katk\u0131 sa\u011flar?<\/p>\n<p>\u00c7oklu i\u015f par\u00e7ac\u0131\u011f\u0131na ihtiya\u00e7 duyulan ba\u015fl\u0131ca senaryolar, genellikle y\u00fcksek hesaplama g\u00fcc\u00fc gerektiren ve paralel olarak y\u00fcr\u00fct\u00fclebilen g\u00f6revlerdir. \u00d6rne\u011fin, bir e-ticaret platformunda, kullan\u0131c\u0131lar\u0131n y\u00fckledi\u011fi y\u00fcksek \u00e7\u00f6z\u00fcn\u00fcrl\u00fckl\u00fc g\u00f6rsellerin farkl\u0131 boyutlarda yeniden boyutland\u0131r\u0131lmas\u0131 ve optimize edilmesi gerekebilir. Bu i\u015flem, tek bir resim i\u00e7in bile zaman al\u0131c\u0131 olabilirken, y\u00fczlerce veya binlerce resim i\u00e7in ana i\u015f par\u00e7ac\u0131\u011f\u0131nda yap\u0131lmas\u0131 imkans\u0131z hale gelir. Benzer \u015fekilde, b\u00fcy\u00fck veri k\u00fcmeleri \u00fczerinde karma\u015f\u0131k analizler yapmak, makine \u00f6\u011frenimi modellerini e\u011fitmek veya \u00e7al\u0131\u015ft\u0131rmak, yo\u011fun \u015fifreleme\/\u015fifre \u00e7\u00f6zme i\u015flemleri ger\u00e7ekle\u015ftirmek, video i\u015fleme (transcoding) veya PDF olu\u015fturma gibi g\u00f6revler, CPU’yu yo\u011fun bir \u015fekilde kullan\u0131r. Bu t\u00fcr g\u00f6revleri ayr\u0131 i\u015f par\u00e7ac\u0131klar\u0131na delege ederek, ana Node.js uygulamas\u0131n\u0131n yan\u0131t verme yetene\u011fini koruyabilir ve kullan\u0131c\u0131 deneyimini iyile\u015ftirebiliriz.<\/p>\n<div class=\"case-study\">\n<h3>Vaka Analizi 1: B\u00fcy\u00fck Bir JSON Dosyas\u0131n\u0131 Paralel Olarak \u0130\u015flemek<\/h3>\n<p>Bir uygulaman\u0131n, y\u00fczlerce megabayt hatta gigabayt boyutunda, i\u00e7 i\u00e7e ge\u00e7mi\u015f nesneler i\u00e7eren b\u00fcy\u00fck bir JSON dosyas\u0131n\u0131 i\u015flemesi gerekti\u011fini d\u00fc\u015f\u00fcn\u00fcn. Bu dosyan\u0131n okunmas\u0131, ayr\u0131\u015ft\u0131r\u0131lmas\u0131 ve i\u00e7erisindeki her bir kayd\u0131n belirli bir mant\u0131\u011fa g\u00f6re d\u00f6n\u00fc\u015ft\u00fcr\u00fclerek veritaban\u0131na kaydedilmesi gerekebilir. Tek i\u015f par\u00e7ac\u0131kl\u0131 bir Node.js uygulamas\u0131nda, bu i\u015flem Event Loop’u uzun s\u00fcre bloke edecektir. Kullan\u0131c\u0131lar bu s\u0131rada yeni istek g\u00f6nderemeyecek veya mevcut oturumlar\u0131 donacakt\u0131r. \u00c7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 kullanarak, bu b\u00fcy\u00fck JSON dosyas\u0131n\u0131 mant\u0131ksal par\u00e7alara ay\u0131rabiliriz. Her bir par\u00e7ay\u0131 ayr\u0131 bir Worker Thread’e g\u00f6ndererek paralel olarak ayr\u0131\u015ft\u0131rma ve i\u015fleme yetene\u011fi kazan\u0131r\u0131z. Bu, toplam i\u015fleme s\u00fcresini \u00f6nemli \u00f6l\u00e7\u00fcde k\u0131salt\u0131r ve ana uygulaman\u0131n kesintisiz \u00e7al\u0131\u015fmas\u0131n\u0131 sa\u011flar.<\/p>\n<\/div>\n<div class=\"case-study\">\n<h3>Vaka Analizi 2: Sunucuda E\u015f Zamanl\u0131 Olarak Birden Fazla Karma\u015f\u0131k Rapor Olu\u015fturma<\/h3>\n<p>Kurumsal uygulamalarda, y\u00f6neticilerin anl\u0131k olarak karma\u015f\u0131k finansal raporlar, sat\u0131\u015f analizleri veya operasyonel \u00f6zetler talep etmesi yayg\u0131nd\u0131r. Bu raporlar genellikle b\u00fcy\u00fck veritaban\u0131 sorgular\u0131n\u0131, veri birle\u015ftirme, hesaplamalar ve PDF veya Excel format\u0131na d\u00f6n\u00fc\u015ft\u00fcrme gibi CPU yo\u011fun ad\u0131mlar\u0131 i\u00e7erir. E\u011fer birden fazla y\u00f6netici ayn\u0131 anda rapor talebinde bulunursa ve bu i\u015flemler ana i\u015f par\u00e7ac\u0131\u011f\u0131nda y\u00fcr\u00fct\u00fcl\u00fcrse, her bir raporun tamamlanma s\u00fcresi katlanarak artar. Worker Threads kullanarak, her bir rapor olu\u015fturma talebini ayr\u0131 bir Worker Thread’e delege edebiliriz. Bu sayede, her rapor paralel olarak i\u015flenir, ana sunucu i\u015f par\u00e7ac\u0131\u011f\u0131 serbest kal\u0131r ve raporlar \u00e7ok daha h\u0131zl\u0131 bir \u015fekilde kullan\u0131c\u0131lara sunulabilir. Bu yakla\u015f\u0131m, \u00f6zellikle yo\u011fun kullan\u0131ml\u0131 sistemlerde \u00f6l\u00e7eklenebilirlik ve yan\u0131t verme yetene\u011fi a\u00e7\u0131s\u0131ndan kritik \u00f6neme sahiptir.<\/p>\n<\/div>\n<p>\u00d6zetle, Node.js’te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131na duyulan ihtiya\u00e7, \u00f6zellikle CPU-bound g\u00f6revlerin uygulaman\u0131n genel performans\u0131n\u0131 ve yan\u0131t verme yetene\u011fini d\u00fc\u015f\u00fcrmesini engellemekten kaynaklan\u0131r. Worker Threads mod\u00fcl\u00fc, bu t\u00fcr g\u00f6revleri ana i\u015f par\u00e7ac\u0131\u011f\u0131ndan ay\u0131rarak, Node.js uygulamalar\u0131n\u0131n daha geni\u015f bir yelpazedeki kullan\u0131m senaryolar\u0131nda y\u00fcksek performans ve \u00f6l\u00e7eklenebilirlik sunmas\u0131na olanak tan\u0131r. Bu, Node.js’in sadece I\/O yo\u011fun uygulamalar i\u00e7in de\u011fil, ayn\u0131 zamanda hesaplama yo\u011fun i\u015f y\u00fckleri i\u00e7in de g\u00fc\u00e7l\u00fc bir se\u00e7enek haline gelmesini sa\u011flar.<\/p>\n<h2>Node.js Worker Threads Mod\u00fcl\u00fc Nedir ve Nas\u0131l \u00c7al\u0131\u015f\u0131r?<\/h2>\n<p>Node.js’in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi k\u0131s\u0131tlamalar\u0131 a\u015fmak i\u00e7in geli\u015ftirilen en \u00f6nemli mod\u00fcllerden biri, \u015f\u00fcphesiz Worker Threads mod\u00fcl\u00fcd\u00fcr. Bu mod\u00fcl, Node.js uygulamalar\u0131na ger\u00e7ek \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 yetene\u011fi kazand\u0131rarak, CPU yo\u011fun g\u00f6revlerin ana i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 bloklamadan paralel olarak y\u00fcr\u00fct\u00fclmesine olanak tan\u0131r. Worker Threads, Node.js v10.5.0 s\u00fcr\u00fcm\u00fcnde deneysel olarak tan\u0131t\u0131lm\u0131\u015f ve v12.11.0 s\u00fcr\u00fcm\u00fcnden itibaren kararl\u0131 hale gelmi\u015ftir. Peki, bu mod\u00fcl tam olarak nedir ve uygulamalar\u0131m\u0131zda nas\u0131l bir i\u015fleyi\u015fe sahiptir?<\/p>\n<p>Worker Threads, temel olarak, ana Node.js s\u00fcrecinin (main thread) yan\u0131nda ayr\u0131 JavaScript y\u00fcr\u00fctme ortamlar\u0131 (worker threads) olu\u015fturman\u0131z\u0131 sa\u011flar. Her bir worker thread, kendi Event Loop’una, kendi haf\u0131za alan\u0131na (ancak SharedArrayBuffer ile payla\u015f\u0131ml\u0131 haf\u0131za da m\u00fcmk\u00fcnd\u00fcr) ve kendi V8 JavaScript motoru \u00f6rne\u011fine sahiptir. Bu, worker’lar\u0131n ana i\u015f par\u00e7ac\u0131\u011f\u0131ndan tamamen ba\u011f\u0131ms\u0131z bir \u015fekilde \u00e7al\u0131\u015fabilece\u011fi anlam\u0131na gelir. Ana i\u015f par\u00e7ac\u0131\u011f\u0131, bir worker thread olu\u015fturdu\u011funda, bu worker’a bir g\u00f6rev atar ve ard\u0131ndan kendi i\u015fine geri d\u00f6ner. Worker, atanan g\u00f6revi kendi ortam\u0131nda paralel olarak y\u00fcr\u00fct\u00fcr ve i\u015fi bitti\u011finde sonucu ana i\u015f par\u00e7ac\u0131\u011f\u0131na geri g\u00f6nderir. Bu ileti\u015fim, bir mesajla\u015fma mekanizmas\u0131 arac\u0131l\u0131\u011f\u0131yla ger\u00e7ekle\u015fir.<\/p>\n<p>Ana i\u015f par\u00e7ac\u0131\u011f\u0131 ile worker i\u015f par\u00e7ac\u0131\u011f\u0131 aras\u0131ndaki ili\u015fki, bir patron-\u00e7al\u0131\u015fan ili\u015fkisine benzetilebilir. Patron (ana i\u015f par\u00e7ac\u0131\u011f\u0131), \u00e7al\u0131\u015fanlara (worker threads) i\u015f da\u011f\u0131t\u0131r ve i\u015fin tamamlanmas\u0131n\u0131 beklerken kendi ana i\u015flerine devam eder. \u00c7al\u0131\u015fanlar ise kendilerine verilen i\u015fi ba\u011f\u0131ms\u0131z olarak yapar ve i\u015f bitti\u011finde patrona sonucu bildirir. Bu ileti\u015fim, <code class=\"language-\">postMessage()<\/code> ve <code class=\"language-\">on('message')<\/code> olay dinleyicileri arac\u0131l\u0131\u011f\u0131yla sa\u011flan\u0131r. <code class=\"language-\">postMessage()<\/code> fonksiyonu, bir i\u015f par\u00e7ac\u0131\u011f\u0131ndan di\u011ferine veri g\u00f6ndermek i\u00e7in kullan\u0131l\u0131rken, <code class=\"language-\">on('message')<\/code> ise gelen veriyi dinlemek i\u00e7in kullan\u0131l\u0131r. G\u00f6nderilen veriler, yap\u0131sal klonlama (structured cloning) algoritmas\u0131 kullan\u0131larak kopyalan\u0131r, yani bellekte ayr\u0131 birer kopyas\u0131 olu\u015fturulur. Bu, worker’lar\u0131n birbirlerinin belle\u011fine do\u011frudan eri\u015fmesini engeller ve veri b\u00fct\u00fcnl\u00fc\u011f\u00fcn\u00fc korur.<\/p>\n<p>Daha ileri d\u00fczey senaryolarda, worker’lar aras\u0131nda payla\u015f\u0131ml\u0131 bellek (SharedArrayBuffer) kullan\u0131m\u0131 da m\u00fcmk\u00fcnd\u00fcr. <code class=\"language-\">SharedArrayBuffer<\/code>, birden fazla worker thread’in ayn\u0131 bellek blo\u011funa eri\u015fmesine olanak tan\u0131r. Bu, b\u00fcy\u00fck veri k\u00fcmelerini kopyalamak yerine referansla payla\u015farak veri transferi maliyetini d\u00fc\u015f\u00fcr\u00fcr. Ancak, payla\u015f\u0131ml\u0131 bellek kullan\u0131rken veri yar\u0131\u015flar\u0131n\u0131 (race conditions) ve tutars\u0131zl\u0131klar\u0131 \u00f6nlemek i\u00e7in <code class=\"language-\">Atomics<\/code> nesnesi gibi senkronizasyon ara\u00e7lar\u0131 kullanmak kritik \u00f6neme sahiptir. <code class=\"language-\">Atomics<\/code> operasyonlar\u0131, payla\u015f\u0131ml\u0131 bellekteki de\u011ferler \u00fczerinde atomik (b\u00f6l\u00fcnemez) i\u015flemler yaparak bu t\u00fcr sorunlar\u0131 engeller.<\/p>\n<h3>Ad\u0131m Ad\u0131m Worker Thread Kullan\u0131m\u0131: Basit Bir \u00d6rnek<\/h3>\n<p>\u015eimdi, Worker Threads mod\u00fcl\u00fcn\u00fc kullanarak basit bir CPU yo\u011fun i\u015flemi paralel hale getiren bir \u00f6rnek \u00fczerinden konuyu daha iyi anlayal\u0131m. Bu \u00f6rnekte, ana i\u015f par\u00e7ac\u0131\u011f\u0131 bir say\u0131 hesaplama g\u00f6revini bir worker’a devredecek ve sonucu geri alacakt\u0131r.<\/p>\n<p>\u00d6ncelikle, worker’\u0131n \u00e7al\u0131\u015ft\u0131raca\u011f\u0131 kodu i\u00e7eren bir JavaScript dosyas\u0131 olu\u015ftural\u0131m. Bu dosyaya \u00f6rne\u011fin <code>worker.js<\/code> ad\u0131n\u0131 verelim:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ worker.js\nconst { parentPort } = require('worker_threads');\n\n\/\/ Ana i\u015f par\u00e7ac\u0131\u011f\u0131ndan gelen mesaj\u0131 dinle\nparentPort.on('message', (message) => {\n if (message.type === 'calculate') {\n const { number } = message;\n let result = 0;\n \/\/ Basit bir CPU yo\u011fun hesaplama \u00f6rne\u011fi\n for (let i = 0; i < number; i++) {\n result += Math.sqrt(i) * Math.sin(i);\n }\n \/\/ Sonucu ana i\u015f par\u00e7ac\u0131\u011f\u0131na geri g\u00f6nder\n parentPort.postMessage({ type: 'result', data: result });\n }\n});\n\nparentPort.postMessage({ type: 'ready' }); \/\/ Worker'\u0131n haz\u0131r oldu\u011funu bildir\n<\/code><\/pre>\n<p>Yukar\u0131daki kodda, <code>parentPort<\/code> nesnesi, worker'\u0131n ana i\u015f par\u00e7ac\u0131\u011f\u0131 ile ileti\u015fim kurmas\u0131n\u0131 sa\u011flayan bir arabirimdir. <code>on('message')<\/code> ile ana i\u015f par\u00e7ac\u0131\u011f\u0131ndan gelen mesajlar\u0131 dinleriz. Mesaj\u0131n <code>type<\/code>'\u0131 'calculate' ise, gelen say\u0131y\u0131 kullanarak basit bir d\u00f6ng\u00fc ile CPU yo\u011fun bir hesaplama yapar\u0131z. Hesaplama tamamland\u0131\u011f\u0131nda, <code>postMessage()<\/code> ile sonucu ana i\u015f par\u00e7ac\u0131\u011f\u0131na geri g\u00f6ndeririz. Ayr\u0131ca, worker'\u0131n ba\u015flat\u0131ld\u0131ktan sonra haz\u0131r oldu\u011funu bildiren bir 'ready' mesaj\u0131 g\u00f6nderiyoruz.<\/p>\n<p>\u015eimdi de bu worker'\u0131 ba\u015flatacak ve onunla ileti\u015fim kuracak olan ana Node.js dosyam\u0131z\u0131 (\u00f6rne\u011fin <code>main.js<\/code>) olu\u015ftural\u0131m:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ main.js\nconst { Worker } = require('worker_threads');\n\nconsole.log('Ana i\u015f par\u00e7ac\u0131\u011f\u0131 ba\u015flad\u0131.');\n\n\/\/ Worker'\u0131 ba\u015flat\nconst worker = new Worker('.\/worker.js');\n\n\/\/ Worker'dan gelen mesajlar\u0131 dinle\nworker.on('message', (message) => {\n if (message.type === 'ready') {\n console.log('Worker haz\u0131r, hesaplama g\u00f6revi g\u00f6nderiliyor...');\n \/\/ Worker'a hesaplama g\u00f6revi g\u00f6nder\n worker.postMessage({ type: 'calculate', number: 20000000 });\n } else if (message.type === 'result') {\n console.log(<code class=\"language-\">Worker'dan gelen sonu\u00e7: ${message.data}<\/code>);\n worker.terminate(); \/\/ Worker'\u0131 sonland\u0131r\n console.log('Ana i\u015f par\u00e7ac\u0131\u011f\u0131 devam ediyor.');\n }\n});\n\n\/\/ Worker'da olu\u015fabilecek hatalar\u0131 yakala\nworker.on('error', (err) => {\n console.error('Worker hatas\u0131:', err);\n});\n\n\/\/ Worker sonland\u0131r\u0131ld\u0131\u011f\u0131nda tetiklenir\nworker.on('exit', (code) => {\n if (code !== 0)\n console.error(<code class=\"language-\">Worker beklenmedik \u015fekilde sonland\u0131. \u00c7\u0131k\u0131\u015f kodu: ${code}<\/code>);\n else\n console.log('Worker ba\u015far\u0131yla sonland\u0131r\u0131ld\u0131.');\n});\n\nconsole.log('Ana i\u015f par\u00e7ac\u0131\u011f\u0131 di\u011fer i\u015flere devam edebilir...');\n<\/code><\/pre>\n<p>Bu <code>main.js<\/code> dosyas\u0131nda, <code>Worker<\/code> s\u0131n\u0131f\u0131n\u0131 kullanarak <code>worker.js<\/code> dosyam\u0131z\u0131 yeni bir i\u015f par\u00e7ac\u0131\u011f\u0131nda ba\u015flat\u0131yoruz. <code>worker.on('message')<\/code> ile worker'dan gelen mesajlar\u0131 dinliyoruz. Worker haz\u0131r oldu\u011funda, ona <code>calculate<\/code> tipinde bir mesaj ve hesaplanacak say\u0131y\u0131 g\u00f6nderiyoruz. Worker'dan sonu\u00e7 geldi\u011finde ise bunu konsola yazd\u0131r\u0131p <code>worker.terminate()<\/code> ile worker'\u0131 sonland\u0131r\u0131yoruz. Ayr\u0131ca, <code>error<\/code> ve <code>exit<\/code> olay dinleyicileri ile hata y\u00f6netimi ve worker'\u0131n ya\u015fam d\u00f6ng\u00fcs\u00fc takibini sa\u011fl\u0131yoruz.<\/p>\n<p>Bu basit \u00f6rnek, Worker Threads'in temel kullan\u0131m\u0131n\u0131 g\u00f6stermektedir. Ana i\u015f par\u00e7ac\u0131\u011f\u0131, yo\u011fun hesaplamay\u0131 worker'a delege ederek kendi i\u015fine devam edebilir. Bu, Node.js uygulamalar\u0131nda performans\u0131 art\u0131rmak ve yan\u0131t verme yetene\u011fini korumak i\u00e7in g\u00fc\u00e7l\u00fc bir y\u00f6ntemdir. \u00d6zellikle CPU yo\u011fun g\u00f6revlerin uygulaman\u0131n genel performans\u0131n\u0131 d\u00fc\u015f\u00fcrmesini engellemek i\u00e7in kritik bir rol oynar. Bu sayede, kullan\u0131c\u0131lar uygulaman\u0131n kesintisiz ve h\u0131zl\u0131 bir \u015fekilde \u00e7al\u0131\u015fmaya devam etti\u011fini deneyimlerler.<\/p>\n<h2>Ger\u00e7ek D\u00fcnya Uygulamalar\u0131 \u0130\u00e7in Worker Threads: Vaka Analizleri ve Uygulamal\u0131 \u00d6rnekler<\/h2>\n<p>Worker Threads mod\u00fcl\u00fcn\u00fcn temel \u00e7al\u0131\u015fma prensibini ve basit bir kullan\u0131m \u00f6rne\u011fini g\u00f6rd\u00fck. \u015eimdi ise bu bilgiyi daha karma\u015f\u0131k, ger\u00e7ek d\u00fcnya senaryolar\u0131na uygulayarak, Node.js uygulamalar\u0131n\u0131zda performans\u0131 nas\u0131l d\u00f6n\u00fc\u015ft\u00fcrebilece\u011finizi inceleyelim. \u00d6zellikle b\u00fcy\u00fck veri i\u015fleme ve kriptografik i\u015flemler gibi CPU yo\u011fun alanlarda Worker Threads'in sa\u011flad\u0131\u011f\u0131 avantajlar olduk\u00e7a b\u00fcy\u00fckt\u00fcr.<\/p>\n<h3>B\u00fcy\u00fck Veri \u0130\u015fleme: JSON Dosyalar\u0131n\u0131 Paralel Olarak Ayr\u0131\u015ft\u0131rma<\/h3>\n<p>G\u00fcn\u00fcm\u00fcz uygulamalar\u0131nda, API'lar arac\u0131l\u0131\u011f\u0131yla veya dosya sisteminden al\u0131nan b\u00fcy\u00fck JSON veri k\u00fcmelerini i\u015flemek s\u0131k\u00e7a kar\u015f\u0131la\u015f\u0131lan bir durumdur. Bu t\u00fcr dosyalar bazen y\u00fczlerce megabayt veya gigabayt boyutunda olabilir ve bunlar\u0131 tek i\u015f par\u00e7ac\u0131\u011f\u0131nda ayr\u0131\u015ft\u0131rmak (parse etmek) ve i\u015flemek, uygulaman\u0131n donmas\u0131na neden olabilir. Worker Threads, bu sorunu \u00e7\u00f6zmek i\u00e7in ideal bir ara\u00e7t\u0131r.<\/p>\n<div class=\"case-study\">\n<h4>Senaryo: \u00c7ok B\u00fcy\u00fck Bir JSON Dosyas\u0131n\u0131 Par\u00e7alara Ay\u0131r\u0131p Her Par\u00e7ay\u0131 Ayr\u0131 Bir Worker'da \u0130\u015flemek<\/h4>\n<p>Diyelim ki, her biri binlerce m\u00fc\u015fteri kayd\u0131 i\u00e7eren 1 GB boyutunda bir JSON dosyam\u0131z var ve bu kay\u0131tlardaki belirli alanlar\u0131 d\u00f6n\u00fc\u015ft\u00fcr\u00fcp, baz\u0131lar\u0131n\u0131 filtreleyip veritaban\u0131na kaydetmemiz gerekiyor. Tek bir Node.js s\u00fcreci bu i\u015flemi yaparken \u00e7ok uzun s\u00fcrebilir ve di\u011fer istekleri bloklayabilir. \u00c7\u00f6z\u00fcm olarak, dosyay\u0131 mant\u0131ksal olarak k\u00fc\u00e7\u00fck par\u00e7alara b\u00f6lebilir ve her bir par\u00e7ay\u0131 ayr\u0131 bir Worker Thread'e g\u00f6ndererek paralel olarak i\u015fleyebiliriz. Sonu\u00e7lar daha sonra ana i\u015f par\u00e7ac\u0131\u011f\u0131nda birle\u015ftirilebilir.<\/p>\n<\/div>\n<p>Bu senaryoyu ger\u00e7ekle\u015ftirmek i\u00e7in iki ana bile\u015fene ihtiyac\u0131m\u0131z var: ana i\u015flemci (main process) ve worker'lar. Ana i\u015flemci dosyay\u0131 okuyacak, par\u00e7alara ay\u0131racak ve worker'lar\u0131 y\u00f6netecektir. Worker'lar ise kendilerine atanan JSON par\u00e7as\u0131n\u0131 ayr\u0131\u015ft\u0131r\u0131p i\u015fleyecektir.<\/p>\n<p>\u0130lk olarak, <code>json-worker.js<\/code> ad\u0131nda worker dosyam\u0131z\u0131 olu\u015ftural\u0131m:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ json-worker.js\nconst { parentPort } = require('worker_threads');\n\nparentPort.on('message', (message) => {\n if (message.type === 'process_chunk') {\n const { chunkId, jsonData } = message;\n console.log(<code class=\"language-\">Worker ${chunkId} par\u00e7ay\u0131 i\u015flemeye ba\u015flad\u0131.<\/code>);\n try {\n \/\/ JSON verisini ayr\u0131\u015ft\u0131r\n const parsedData = JSON.parse(jsonData);\n\n \/\/ Burada veri \u00fczerinde CPU yo\u011fun i\u015flemler yap\u0131labilir\n \/\/ \u00d6rne\u011fin: Her bir kayd\u0131 d\u00f6n\u00fc\u015ft\u00fcrme, filtreleme, karma\u015f\u0131k hesaplamalar\n const processedData = parsedData.map(record => ({\n id: record.id,\n name: record.name.toUpperCase(), \/\/ \u00d6rnek d\u00f6n\u00fc\u015f\u00fcm\n age: record.age + 1 \/\/ \u00d6rnek hesaplama\n })).filter(record => record.age > 20); \/\/ \u00d6rnek filtreleme\n\n parentPort.postMessage({ type: 'chunk_processed', chunkId, result: processedData });\n } catch (error) {\n parentPort.postMessage({ type: 'error', chunkId, error: error.message });\n }\n }\n});\n<\/code><\/pre>\n<p>\u015eimdi de <code>main.js<\/code> dosyam\u0131z\u0131 olu\u015ftural\u0131m. Bu dosya, b\u00fcy\u00fck bir sahte JSON dosyas\u0131n\u0131 okuyacak ve onu birden fazla worker'a da\u011f\u0131tacakt\u0131r:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ main.js\nconst { Worker } = require('worker_threads');\nconst fs = require('fs');\nconst path = require('path');\n\nconst NUM_WORKERS = 4; \/\/ Ka\u00e7 worker kullanaca\u011f\u0131m\u0131z\u0131 belirleyelim\nconst CHUNK_SIZE = 1024 * 1024; \/\/ 1MB'l\u0131k par\u00e7alar halinde okuyal\u0131m\n\nasync function processLargeJsonFile(filePath) {\n console.log('B\u00fcy\u00fck JSON dosyas\u0131n\u0131 i\u015fleme ba\u015flad\u0131...');\n const workers = [];\n const results = [];\n let processedChunks = 0;\n let totalChunks = 0;\n\n \/\/ Worker'lar\u0131 ba\u015flat\n for (let i = 0; i < NUM_WORKERS; i++) {\n const worker = new Worker(path.join(__dirname, 'json-worker.js'));\n workers.push(worker);\n\n worker.on('message', (message) => {\n if (message.type === 'chunk_processed') {\n results.push(...message.result);\n processedChunks++;\n console.log(<code class=\"language-\">Par\u00e7a ${message.chunkId} i\u015flendi. Toplam i\u015flenen: ${processedChunks}\/${totalChunks}<\/code>);\n if (processedChunks === totalChunks) {\n console.log('T\u00fcm par\u00e7alar i\u015flendi. Toplam kay\u0131t:', results.length);\n workers.forEach(w => w.terminate()); \/\/ T\u00fcm worker'lar\u0131 sonland\u0131r\n console.log('T\u00fcm veriler ba\u015far\u0131yla i\u015flendi ve birle\u015ftirildi.');\n \/\/ console.log(results.slice(0, 5)); \/\/ \u0130lk 5 kayd\u0131 g\u00f6ster\n }\n } else if (message.type === 'error') {\n console.error(<code class=\"language-\">Worker ${message.chunkId} hatas\u0131:<\/code>, message.error);\n processedChunks++; \/\/ Hatal\u0131 par\u00e7ay\u0131 da say\n if (processedChunks === totalChunks) {\n workers.forEach(w => w.terminate());\n console.log('T\u00fcm par\u00e7alar i\u015flendi (hatal\u0131 olanlar dahil).');\n }\n }\n });\n\n worker.on('error', (err) => console.error('Worker hatas\u0131:', err));\n worker.on('exit', (code) => {\n if (code !== 0) console.error(<code class=\"language-\">Worker exit with code ${code}<\/code>);\n });\n }\n\n \/\/ B\u00fcy\u00fck dosyay\u0131 par\u00e7alar halinde oku ve worker'lara g\u00f6nder\n const fileStream = fs.createReadStream(filePath, { encoding: 'utf8', highWaterMark: CHUNK_SIZE });\n let buffer = '';\n let chunkId = 0;\n\n for await (const chunk of fileStream) {\n buffer += chunk;\n \/\/ JSON'\u0131 par\u00e7alara ay\u0131rmak i\u00e7in basit bir yakla\u015f\u0131m\n \/\/ Ger\u00e7ek bir senaryoda daha robust bir JSON stream parser gerekebilir\n let lastBracketIndex = buffer.lastIndexOf('}');\n if (lastBracketIndex !== -1) {\n let jsonPart = buffer.substring(0, lastBracketIndex + 1);\n buffer = buffer.substring(lastBracketIndex + 1);\n\n \/\/ JSON array'i a\u00e7\u0131p kapatmak i\u00e7in d\u00fczeltme\n if (jsonPart.startsWith('[')) jsonPart = jsonPart.substring(1);\n if (jsonPart.endsWith(']')) jsonPart = jsonPart.substring(0, jsonPart.length - 1);\n \n \/\/ Birden fazla worker'a da\u011f\u0131tmak i\u00e7in\n const workerIndex = chunkId % NUM_WORKERS;\n workers[workerIndex].postMessage({ type: 'process_chunk', chunkId: chunkId++, jsonData: <code class=\"language-\">[${jsonPart}]<\/code> });\n totalChunks++;\n }\n }\n\n \/\/ Kalan buffer'\u0131 i\u015fle (son par\u00e7a i\u00e7in)\n if (buffer.trim().length > 0) {\n let jsonPart = buffer.trim();\n if (jsonPart.startsWith('[')) jsonPart = jsonPart.substring(1);\n if (jsonPart.endsWith(']')) jsonPart = jsonPart.substring(0, jsonPart.length - 1);\n \n const workerIndex = chunkId % NUM_WORKERS;\n workers[workerIndex].postMessage({ type: 'process_chunk', chunkId: chunkId++, jsonData: <code class=\"language-\">[${jsonPart}]<\/code> });\n totalChunks++;\n }\n\n \/\/ T\u00fcm worker'lar\u0131n i\u015fi bitirmesi bekleniyor\n \/\/ Bu k\u0131s\u0131m mesaj dinleyicisi i\u00e7inde y\u00f6netiliyor\n}\n\n\/\/ \u00d6rnek bir sahte b\u00fcy\u00fck JSON dosyas\u0131 olu\u015fturma fonksiyonu\nfunction createDummyLargeJsonFile(filename, numRecords) {\n const records = [];\n for (let i = 0; i < numRecords; i++) {\n records.push({\n id: i,\n name: <code class=\"language-\">User ${i}<\/code>,\n email: <code class=\"language-\">user${i}@example.com<\/code>,\n age: Math.floor(Math.random() * 50) + 18,\n city: <code class=\"language-\">City ${Math.floor(Math.random() * 10)}<\/code>\n });\n }\n fs.writeFileSync(filename, JSON.stringify(records, null, 2));\n console.log(<code class=\"language-\">${filename} dosyas\u0131 olu\u015fturuldu, ${numRecords} kay\u0131t i\u00e7eriyor.<\/code>);\n}\n\nconst dummyFileName = 'large_data.json';\ncreateDummyLargeJsonFile(dummyFileName, 100000); \/\/ 100.000 kay\u0131tl\u0131k dosya olu\u015ftur\n\nprocessLargeJsonFile(dummyFileName).catch(console.error);\n<\/code><\/pre>\n<p>Bu \u00f6rnek, b\u00fcy\u00fck bir JSON dosyas\u0131n\u0131 okuyup, her bir worker'a veri par\u00e7ac\u0131klar\u0131 g\u00f6ndererek paralel i\u015flemeyi g\u00f6stermektedir. Ger\u00e7ek bir senaryoda, JSON par\u00e7alama mant\u0131\u011f\u0131 daha karma\u015f\u0131k olabilir (\u00f6rne\u011fin, bir JSON ak\u0131\u015f ayr\u0131\u015ft\u0131r\u0131c\u0131s\u0131 kullanmak), ancak temel fikir ayn\u0131d\u0131r: CPU yo\u011fun i\u015fi birden fazla worker'a da\u011f\u0131tmak. Bu y\u00f6ntem, tek i\u015f par\u00e7ac\u0131kl\u0131 bir yakla\u015f\u0131ma g\u00f6re i\u015flem s\u00fcresini \u00f6nemli \u00f6l\u00e7\u00fcde h\u0131zland\u0131rabilir ve ana Node.js s\u00fcrecinin yan\u0131t verme yetene\u011fini korur.<\/p>\n<h3>Kriptografik \u0130\u015flemlerin H\u0131zland\u0131r\u0131lmas\u0131<\/h3>\n<p>Kriptografik algoritmalar, \u015fifreleme, \u015fifre \u00e7\u00f6zme, hash hesaplama gibi i\u015flemler genellikle CPU yo\u011fun g\u00f6revlerdir. \u00d6zellikle birden fazla verinin ayn\u0131 anda i\u015flenmesi gerekti\u011finde, Worker Threads bu i\u015flemlerin paralel olarak y\u00fcr\u00fct\u00fclmesini sa\u011flayarak performans\u0131 art\u0131rabilir.<\/p>\n<div class=\"case-study\">\n<h4>Senaryo: Yo\u011fun Hash Hesaplamalar\u0131n\u0131 Worker'lara Da\u011f\u0131tma<\/h4>\n<p>Bir uygulaman\u0131n, kullan\u0131c\u0131 parolalar\u0131n\u0131 veya b\u00fcy\u00fck dosyalar\u0131n b\u00fct\u00fcnl\u00fc\u011f\u00fcn\u00fc do\u011frulamak i\u00e7in SHA256 gibi algoritmalarla yo\u011fun hash hesaplamalar\u0131 yapmas\u0131 gerekti\u011fini d\u00fc\u015f\u00fcn\u00fcn. E\u011fer ayn\u0131 anda birden fazla hash hesaplamas\u0131 tetiklenirse, ana i\u015f par\u00e7ac\u0131\u011f\u0131 kilitlenebilir. Worker Threads kullanarak, her bir hash hesaplama g\u00f6revini ayr\u0131 bir worker'a atayabiliriz.<\/p>\n<\/div>\n<p><code>hash-worker.js<\/code> ad\u0131nda bir worker dosyas\u0131 olu\u015ftural\u0131m:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ hash-worker.js\nconst { parentPort } = require('worker_threads');\nconst crypto = require('crypto');\n\nparentPort.on('message', (message) => {\n if (message.type === 'hash') {\n const { dataToHash, id } = message;\n console.log(<code class=\"language-\">Worker ${id}: Hash hesaplamaya ba\u015flad\u0131...<\/code>);\n const hash = crypto.createHash('sha256').update(dataToHash).digest('hex');\n console.log(<code class=\"language-\">Worker ${id}: Hash hesaplamay\u0131 bitirdi.<\/code>);\n parentPort.postMessage({ type: 'hashed', id, hash });\n }\n});\n<\/code><\/pre>\n<p>\u015eimdi de <code>main-hash.js<\/code> dosyam\u0131z\u0131 olu\u015ftural\u0131m:<\/p>\n<pre><code class=\"language-javascript\">\n\/\/ main-hash.js\nconst { Worker } = require('worker_threads');\n\nconsole.log('Ana i\u015f par\u00e7ac\u0131\u011f\u0131 hash i\u015flemini ba\u015flatt\u0131.');\n\nconst dataList = [\n 'Bu birinci veri par\u00e7as\u0131d\u0131r ve hashlenecektir.',\n '\u0130kinci \u00f6nemli veri par\u00e7as\u0131 da burada.',\n '\u00dc\u00e7\u00fcnc\u00fc ve son veri par\u00e7as\u0131 da i\u015fleniyor.',\n \/\/ Daha fazla veri eklenebilir\n];\n\nconst workers = [];\nlet completedHashes = 0;\n\ndataList.forEach((data, index) => {\n const worker = new Worker('.\/hash-worker.js');\n workers.push(worker);\n\n worker.on('message', (message) => {\n if (message.type === 'hashed') {\n console.log(<code class=\"language-\">Veri ${message.id} i\u00e7in hash: ${message.hash}<\/code>);\n completedHashes++;\n if (completedHashes === dataList.length) {\n console.log('T\u00fcm hash i\u015flemleri tamamland\u0131.');\n workers.forEach(w => w.terminate());\n }\n }\n });\n\n worker.on('error', (err) => console.error(<code class=\"language-\">Worker ${index} hatas\u0131:<\/code>, err));\n worker.on('exit', (code) => {\n if (code !== 0) console.error(<code class=\"language-\">Worker ${index} beklenmedik \u015fekilde sonland\u0131.<\/code>);\n });\n\n worker.postMessage({ type: 'hash', dataToHash: data, id: index });\n});\n\nconsole.log('Ana i\u015f par\u00e7ac\u0131\u011f\u0131 di\u011fer i\u015flere devam edebilirken hashler hesaplan\u0131yor...');\n<\/code><\/pre>\n<p>Bu \u00f6rnekte, <code>dataList<\/code>'teki her bir veri par\u00e7as\u0131 i\u00e7in ayr\u0131 bir worker ba\u015flat\u0131l\u0131r ve hash hesaplama g\u00f6revi bu worker'lara atan\u0131r. Ana i\u015f par\u00e7ac\u0131\u011f\u0131, t\u00fcm worker'lar i\u015flerini bitirene kadar beklemek zorunda kalmaz, kendi i\u015fine devam edebilir. Bu, \u00f6zellikle birden fazla e\u015f zamanl\u0131 kriptografik i\u015flem gerektiren durumlarda uygulaman\u0131n genel yan\u0131t verme h\u0131z\u0131n\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde art\u0131r\u0131r.<\/p>\n<p>Bu vaka analizleri ve \u00f6rnekler, Worker Threads'in Node.js'te CPU yo\u011fun g\u00f6revleri nas\u0131l etkin bir \u015fekilde paralel hale getirebilece\u011fini ve uygulaman\u0131z\u0131n performans\u0131n\u0131 nas\u0131l art\u0131rabilece\u011fini g\u00f6stermektedir. Do\u011fru kullan\u0131ld\u0131\u011f\u0131nda, Worker Threads Node.js'in yeteneklerini \u00f6nemli \u00f6l\u00e7\u00fcde geni\u015fleten g\u00fc\u00e7l\u00fc bir ara\u00e7t\u0131r.<\/p>\n<h2>Worker Threads \u0130le Performans\u0131 Optimize Etme \u0130pu\u00e7lar\u0131 ve P\u00fcf Noktalar\u0131<\/h2>\n<p>Worker Threads, Node.js uygulamalar\u0131n\u0131zda CPU yo\u011fun g\u00f6revlerin performans\u0131n\u0131 art\u0131rmak i\u00e7in g\u00fc\u00e7l\u00fc bir ara\u00e7t\u0131r. Ancak, bu g\u00fcc\u00fc en verimli \u015fekilde kullanmak ve potansiyel tuzaklardan ka\u00e7\u0131nmak i\u00e7in baz\u0131 \u00f6nemli ipu\u00e7lar\u0131 ve p\u00fcf noktalar\u0131 bulunmaktad\u0131r. Performans optimizasyonu, sadece worker kullanmakla kalmaz, ayn\u0131 zamanda worker'lar\u0131n nas\u0131l y\u00f6netildi\u011fi, veri transferinin nas\u0131l yap\u0131ld\u0131\u011f\u0131 ve hata senaryolar\u0131n\u0131n nas\u0131l ele al\u0131nd\u0131\u011f\u0131 ile de yak\u0131ndan ilgilidir.<\/p>\n<h3>Worker Havuzlar\u0131 (Worker Pools) Olu\u015fturma<\/h3>\n<p>Her CPU yo\u011fun g\u00f6rev i\u00e7in yeni bir worker olu\u015fturmak ve sonland\u0131rmak, kendi ba\u015f\u0131na bir maliyettir. Worker'lar\u0131n ba\u015flat\u0131lmas\u0131 ve kapat\u0131lmas\u0131, kaynak t\u00fcketimi ve zaman a\u00e7\u0131s\u0131ndan ek y\u00fck getirir. \u00d6zellikle k\u0131sa \u00f6m\u00fcrl\u00fc ve s\u0131k tekrarlanan g\u00f6revler i\u00e7in bu maliyet, elde edilen performans kazanc\u0131n\u0131 g\u00f6lgede b\u0131rakabilir. Bu sorunu \u00e7\u00f6zmek i\u00e7in \"Worker Havuzlar\u0131\" (Worker Pools) kullanmak en iyi yakla\u015f\u0131md\u0131r. Bir worker havuzu, belirli say\u0131da worker'\u0131 \u00f6nceden ba\u015flat\u0131r ve bu worker'lar\u0131 yeniden kullanarak g\u00f6revleri da\u011f\u0131t\u0131r. G\u00f6rev tamamland\u0131\u011f\u0131nda worker sonland\u0131r\u0131lmaz, bir sonraki g\u00f6revi beklemek \u00fczere havuza geri d\u00f6ner.<\/p>\n<p>Bir worker havuzu implementasyonu, genel olarak \u015fu ad\u0131mlar\u0131 i\u00e7erir:<\/p>\n<ol>\n<li>Belirli say\u0131da worker'\u0131 ba\u015flat\u0131n ve bir kuyrukta bekletin.<\/li>\n<li>Yeni bir g\u00f6rev geldi\u011finde, kuyruktan bo\u015fta olan bir worker'\u0131 al\u0131n.<\/li>\n<li>G\u00f6revi worker'a g\u00f6nderin ve worker'\u0131n i\u015fini bitirmesini bekleyin.<\/li>\n<li>Worker i\u015fini bitirdi\u011finde, onu tekrar bo\u015fta olan worker'lar kuyru\u011funa ekleyin.<\/li>\n<li>E\u011fer t\u00fcm worker'lar me\u015fgul ise, gelen g\u00f6revleri bir g\u00f6rev kuyru\u011funda bekletin ve bo\u015f bir worker oldu\u011funda i\u015fleyin.<\/li>\n<\/ol>\n<p>Bu yakla\u015f\u0131m, worker ba\u015flatma\/sonland\u0131rma maliyetini ortadan kald\u0131r\u0131r ve kaynak kullan\u0131m\u0131n\u0131 daha verimli hale getirir. Piyasada worker havuzu y\u00f6netimi i\u00e7in baz\u0131 k\u00fct\u00fcphaneler (\u00f6rne\u011fin <code>piscina<\/code>) bulunmaktad\u0131r, bu k\u00fct\u00fcphaneleri kullanmak kendi \u00e7\u00f6z\u00fcm\u00fcn\u00fcz\u00fc yazmaktan daha pratik olabilir.<\/p>\n<h3>Veri Transferi Maliyetini Minimize Etme<\/h3>\n<p>Worker Threads aras\u0131nda veri transferi, performans a\u00e7\u0131s\u0131ndan \u00f6nemli bir fakt\u00f6rd\u00fcr. <code class=\"language-\">postMessage()<\/code> ile veri g\u00f6nderdi\u011finizde, varsay\u0131lan olarak veriler yap\u0131sal klonlama (structured cloning) algoritmas\u0131 kullan\u0131larak kopyalan\u0131r. Bu, b\u00fcy\u00fck veri k\u00fcmeleri i\u00e7in ciddi bir performans d\u00fc\u015f\u00fc\u015f\u00fcne neden olabilir, \u00e7\u00fcnk\u00fc verinin hem g\u00f6nderen hem de al\u0131c\u0131 tarafta bellekte ayr\u0131 kopyalar\u0131 olu\u015fturulur. Bu maliyeti minimize etmek i\u00e7in iki temel y\u00f6ntem vard\u0131r:<\/p>\n<ul>\n<li><strong>Transferable Objects (Aktar\u0131labilir Nesneler):<\/strong> Baz\u0131 JavaScript nesneleri (\u00f6rne\u011fin <code>ArrayBuffer<\/code>, <code>MessagePort<\/code>, <code>FileHandle<\/code>), kopyalanmak yerine do\u011frudan aktar\u0131labilir. Bu, verinin sahipli\u011finin bir i\u015f par\u00e7ac\u0131\u011f\u0131ndan di\u011ferine devredilmesi anlam\u0131na gelir. Aktar\u0131lan veri, g\u00f6nderen i\u015f par\u00e7ac\u0131\u011f\u0131nda art\u0131k kullan\u0131lamaz hale gelir. Bu y\u00f6ntem, b\u00fcy\u00fck ikili veri bloklar\u0131 i\u00e7in idealdir.<\/li>\n<li><strong>SharedArrayBuffer ve Atomics:<\/strong> E\u011fer verinin hem ana i\u015f par\u00e7ac\u0131\u011f\u0131nda hem de worker'da e\u015f zamanl\u0131 olarak eri\u015filmesi ve de\u011fi\u015ftirilmesi gerekiyorsa, <code>SharedArrayBuffer<\/code> kullanmak en uygun \u00e7\u00f6z\u00fcmd\u00fcr. <code>SharedArrayBuffer<\/code>, birden fazla i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n ayn\u0131 bellek blo\u011funu payla\u015fmas\u0131na olanak tan\u0131r. Ancak, bu durumda veri yar\u0131\u015flar\u0131n\u0131 (race conditions) ve tutars\u0131zl\u0131klar\u0131 \u00f6nlemek i\u00e7in <code>Atomics<\/code> nesnesi ile dikkatli senkronizasyon sa\u011flamak zorunludur. <code>Atomics<\/code>, payla\u015f\u0131ml\u0131 bellekteki de\u011ferler \u00fczerinde atomik (b\u00f6l\u00fcnemez) operasyonlar yaparak g\u00fcvenli eri\u015fimi garanti eder. Bu, daha karma\u015f\u0131k bir kullan\u0131m gerektirse de, en y\u00fcksek performansl\u0131 veri payla\u015f\u0131m y\u00f6ntemidir.<\/li>\n<\/ul>\n<div class=\"expert-tip\">\n Uzman \u0130pucu: B\u00fcy\u00fck veri k\u00fcmelerini worker'lara g\u00f6nderirken <code>ArrayBuffer<\/code> kullan\u0131yorsan\u0131z, <code>postMessage(data, [data.buffer])<\/code> \u015feklinde ikinci arg\u00fcman olarak aktar\u0131lacak buffer'\u0131 belirtmeyi unutmay\u0131n. Bu, verinin kopyalanmas\u0131 yerine aktar\u0131lmas\u0131n\u0131 sa\u011flar ve performans\u0131 %40'tan fazla art\u0131rabilir!\n<\/div>\n<h3>Hata Y\u00f6netimi ve Graceful Shutdown<\/h3>\n<p>Worker Threads kullan\u0131rken sa\u011flam bir hata y\u00f6netimi stratejisi geli\u015ftirmek kritik \u00f6neme sahiptir. Bir worker'da olu\u015fan hata, ana i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 do\u011frudan etkilemez ancak i\u015fin tamamlanamamas\u0131na neden olabilir. Worker'larda olu\u015fan hatalar\u0131 yakalamak i\u00e7in <code>worker.on('error', ...)<\/code> ve worker'\u0131n beklenmedik bir \u015fekilde sonlanmas\u0131 durumunda <code>worker.on('exit', ...)<\/code> olay dinleyicilerini kullanmak \u00f6nemlidir. Ayr\u0131ca, ana uygulaman\u0131z kapan\u0131rken t\u00fcm aktif worker'lar\u0131n d\u00fczg\u00fcn bir \u015fekilde sonland\u0131r\u0131ld\u0131\u011f\u0131ndan (graceful shutdown) emin olmal\u0131s\u0131n\u0131z. Aksi takdirde, arka planda \u00e7al\u0131\u015fan worker'lar kaynaklar\u0131 t\u00fcketmeye devam edebilir.<\/p>\n<ul>\n<li><strong>Hata Yakalama:<\/strong> Worker i\u00e7indeki kodda <code>try...catch<\/code> bloklar\u0131 kullanarak olas\u0131 hatalar\u0131 yakalay\u0131n ve <code>parentPort.postMessage({ type: 'error', message: err.message })<\/code> ile ana i\u015f par\u00e7ac\u0131\u011f\u0131na bildirin.<\/li>\n<li><strong>Worker Sonland\u0131rma:<\/strong> \u0130\u015flem tamamland\u0131\u011f\u0131nda veya bir hata olu\u015ftu\u011funda <code>worker.terminate()<\/code> \u00e7a\u011f\u0131rarak worker'\u0131 manuel olarak sonland\u0131r\u0131n. Worker havuzlar\u0131 kullan\u0131yorsan\u0131z, worker'\u0131 havuza geri d\u00f6nd\u00fcrmeden \u00f6nce durumunu kontrol edin.<\/li>\n<\/ul>\n<h3>Worker'lar\u0131n Ya\u015fam D\u00f6ng\u00fcs\u00fc Y\u00f6netimi<\/h3>\n<p>Worker'lar\u0131n ya\u015fam d\u00f6ng\u00fcs\u00fcn\u00fc do\u011fru y\u00f6netmek, uygulaman\u0131z\u0131n kararl\u0131l\u0131\u011f\u0131 ve performans i\u00e7in \u00f6nemlidir. Gereksiz worker'lar\u0131 a\u00e7\u0131k tutmak bellek ve CPU kaynaklar\u0131n\u0131 bo\u015fa harcayabilir. Worker'lar\u0131n ne zaman ba\u015flat\u0131laca\u011f\u0131, ne zaman kullan\u0131laca\u011f\u0131 ve ne zaman sonland\u0131r\u0131laca\u011f\u0131 iyi planlanmal\u0131d\u0131r. Worker havuzlar\u0131, bu y\u00f6netimi otomatikle\u015ftirmenin en etkili yoludur. Ayr\u0131ca, worker'lar\u0131n ba\u015flat\u0131l\u0131rken ba\u011f\u0131ml\u0131l\u0131klar\u0131n\u0131 (mod\u00fcller, veritaban\u0131 ba\u011flant\u0131lar\u0131 vb.) nas\u0131l y\u00f6netece\u011finizi de d\u00fc\u015f\u00fcnmelisiniz. Her worker'\u0131n kendi ba\u011f\u0131ms\u0131z ortam\u0131 oldu\u011fu i\u00e7in, bu ba\u011f\u0131ml\u0131l\u0131klar\u0131n her worker i\u00e7in do\u011fru \u015fekilde y\u00fcklendi\u011finden ve yap\u0131land\u0131r\u0131ld\u0131\u011f\u0131ndan emin olmal\u0131s\u0131n\u0131z.<\/p>\n<p>Bu ipu\u00e7lar\u0131 ve p\u00fcf noktalar\u0131, Node.js Worker Threads'i kullan\u0131rken daha performansl\u0131, kararl\u0131 ve \u00f6l\u00e7eklenebilir uygulamalar geli\u015ftirmenize yard\u0131mc\u0131 olacakt\u0131r. Unutmay\u0131n ki her zaman oldu\u011fu gibi, en iyi performans\u0131 elde etmek i\u00e7in uygulaman\u0131z\u0131n \u00f6zel ihtiya\u00e7lar\u0131na g\u00f6re profil olu\u015fturma ve test etme yapmak esast\u0131r.<\/p>\n<h2>Sonu\u00e7: Node.js'te \u00c7oklu \u0130\u015f Par\u00e7ac\u0131\u011f\u0131n\u0131n Gelece\u011fi ve S\u0131k\u00e7a Sorulan Sorular<\/h2>\n<p>Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 do\u011fas\u0131, \u00f6zellikle I\/O yo\u011fun uygulamalar i\u00e7in e\u015fsiz avantajlar sunarken, CPU yo\u011fun g\u00f6revlerde performans darbo\u011fazlar\u0131na yol a\u00e7abiliyordu. Ancak, Worker Threads mod\u00fcl\u00fcn\u00fcn ortaya \u00e7\u0131k\u0131\u015f\u0131, bu paradigmay\u0131 k\u00f6kten de\u011fi\u015ftirerek Node.js geli\u015ftiricilerine ger\u00e7ek \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 yetenekleri kazand\u0131rm\u0131\u015ft\u0131r. Bu kapsaml\u0131 rehber boyunca, Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n temellerinden ba\u015flayarak, \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131na neden ihtiya\u00e7 duydu\u011fumuzu, Worker Threads mod\u00fcl\u00fcn\u00fcn nas\u0131l \u00e7al\u0131\u015ft\u0131\u011f\u0131n\u0131 ve ger\u00e7ek d\u00fcnya senaryolar\u0131nda nas\u0131l kullan\u0131labilece\u011fini ad\u0131m ad\u0131m inceledik. B\u00fcy\u00fck JSON dosyalar\u0131n\u0131 paralel olarak i\u015flemek ve kriptografik hesaplamalar\u0131 h\u0131zland\u0131rmak gibi vaka analizleriyle, bu teknolojinin pratik uygulamalar\u0131n\u0131 ve potansiyelini g\u00f6zlemledik.<\/p>\n<p>Worker Threads, Node.js'in yeteneklerini geni\u015fleterek, onu sadece I\/O yo\u011fun web sunucular\u0131 i\u00e7in de\u011fil, ayn\u0131 zamanda yo\u011fun hesaplama gerektiren arka plan servisleri, veri analizi ara\u00e7lar\u0131 ve medya i\u015fleme platformlar\u0131 i\u00e7in de g\u00fc\u00e7l\u00fc bir aday haline getirmi\u015ftir. Art\u0131k geli\u015ftiriciler, Node.js ekosisteminin tan\u0131d\u0131k ve verimli yap\u0131s\u0131ndan vazge\u00e7meden, uygulaman\u0131n en kritik ve performans gerektiren k\u0131s\u0131mlar\u0131n\u0131 paralel olarak \u00e7al\u0131\u015ft\u0131rabilirler. Ancak, her g\u00fc\u00e7l\u00fc ara\u00e7ta oldu\u011fu gibi, Worker Threads'in de ne zaman ve nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 iyi anlamak \u00f6nemlidir. Her g\u00f6revi bir worker'a devretmek her zaman en iyi \u00e7\u00f6z\u00fcm de\u011fildir; \u00f6zellikle I\/O yo\u011fun g\u00f6revler i\u00e7in ana i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n non-blocking yap\u0131s\u0131 genellikle yeterlidir ve hatta daha verimlidir. Worker Threads, \u00f6zellikle CPU'yu uzun s\u00fcre me\u015fgul eden ve ana Event Loop'u bloklama potansiyeli olan g\u00f6revler i\u00e7in tasarlanm\u0131\u015ft\u0131r.<\/p>\n<p>Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n gelece\u011fi parlak g\u00f6r\u00fcnmektedir. Topluluk ve \u00e7ekirdek geli\u015ftiriciler, mod\u00fcl\u00fc daha da optimize etmek, kullan\u0131m\u0131n\u0131 kolayla\u015ft\u0131rmak ve yeni \u00f6zellikler eklemek i\u00e7in s\u00fcrekli \u00e7al\u0131\u015fmaktad\u0131r. Worker havuzlar\u0131 i\u00e7in standart \u00e7\u00f6z\u00fcmlerin entegrasyonu, daha geli\u015fmi\u015f hata y\u00f6netimi mekanizmalar\u0131 ve daha esnek veri payla\u015f\u0131m se\u00e7enekleri gibi iyile\u015ftirmeler beklenmektedir. Bu geli\u015fmeler, Node.js'in modern ve karma\u015f\u0131k sistemlerin gereksinimlerini daha iyi kar\u015f\u0131lamas\u0131na olanak tan\u0131yacakt\u0131r. Uygulamalar\u0131n\u0131z\u0131n performans\u0131n\u0131 ve \u00f6l\u00e7eklenebilirli\u011fini bir sonraki seviyeye ta\u015f\u0131mak istiyorsan\u0131z, Worker Threads mod\u00fcl\u00fcn\u00fc \u00f6\u011frenmek ve projelerinize entegre etmek kesinlikle de\u011ferli bir yat\u0131r\u0131md\u0131r.<\/p>\n<h3>S\u0131k\u00e7a Sorulan Sorular (SSS)<\/h3>\n<h4>Soru 1: Worker Threads ile Clustering aras\u0131ndaki fark nedir?<\/h4>\n<p><strong>Cevap:<\/strong> Worker Threads ve Clustering (k\u00fcmeleme) Node.js'te paralellik sa\u011flamak i\u00e7in kullan\u0131lan farkl\u0131 mekanizmalard\u0131r.<\/p>\n<ul>\n<li><strong>Worker Threads:<\/strong> Ayn\u0131 Node.js s\u00fcrecinin i\u00e7inde ayr\u0131 i\u015f par\u00e7ac\u0131klar\u0131 olu\u015fturur. Bu i\u015f par\u00e7ac\u0131klar\u0131, ayn\u0131 bellek alan\u0131n\u0131 (SharedArrayBuffer ile) payla\u015fabilir ve ayn\u0131 uygulaman\u0131n CPU yo\u011fun k\u0131s\u0131mlar\u0131n\u0131 paralel olarak \u00e7al\u0131\u015ft\u0131rmak i\u00e7in idealdir. Genellikle tek bir sunucu \u00fczerindeki CPU \u00e7ekirdeklerini daha verimli kullanmak i\u00e7in kullan\u0131l\u0131r.<\/li>\n<li><strong>Clustering:<\/strong> Node.js'in <code>cluster<\/code> mod\u00fcl\u00fc, ayn\u0131 uygulaman\u0131n birden fazla tam Node.js s\u00fcrecini (child processes) ba\u015flat\u0131r. Her bir s\u00fcre\u00e7 kendi Event Loop'una ve kendi haf\u0131za alan\u0131na sahiptir. Bu s\u00fcre\u00e7ler, ayn\u0131 sunucu portunu payla\u015fabilir ve gelen istekleri da\u011f\u0131tarak I\/O yo\u011fun uygulamalar\u0131n y\u00fck dengelemesini ve daha iyi \u00f6l\u00e7eklenmesini sa\u011flar. Genellikle birden fazla CPU \u00e7ekirde\u011fini kullanarak I\/O yo\u011fun web sunucular\u0131n\u0131n e\u015f zamanl\u0131 istekleri daha verimli i\u015flemesi i\u00e7in kullan\u0131l\u0131r.<\/li>\n<\/ul>\n<p>\u00d6zetle, Worker Threads CPU-bound g\u00f6revler i\u00e7in s\u00fcre\u00e7 i\u00e7i paralellik sa\u011flarken, Clustering I\/O-bound g\u00f6revler i\u00e7in s\u00fcre\u00e7ler aras\u0131 paralellik sa\u011flar.<\/p>\n<h4>Soru 2: Worker Threads her durumda performans\u0131 art\u0131r\u0131r m\u0131?<\/h4>\n<p><strong>Cevap:<\/strong> Hay\u0131r, her durumda performans\u0131 art\u0131rmaz. Worker Threads, \u00f6zellikle CPU-bound (i\u015flemciye ba\u011fl\u0131) g\u00f6revler i\u00e7in performans art\u0131\u015f\u0131 sa\u011flar. E\u011fer uygulaman\u0131z \u00e7o\u011funlukla I\/O-bound (giri\u015f\/\u00e7\u0131k\u0131\u015fa ba\u011fl\u0131) ise (\u00f6rne\u011fin, veritaban\u0131 sorgular\u0131, a\u011f istekleri), Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131, non-blocking I\/O modeli zaten olduk\u00e7a verimlidir ve Worker Threads kullanman\u0131n ek y\u00fck\u00fc (i\u015f par\u00e7ac\u0131\u011f\u0131 olu\u015fturma, veri transferi) performans\u0131 d\u00fc\u015f\u00fcrebilir. Worker Threads kullanmadan \u00f6nce, uygulaman\u0131z\u0131n darbo\u011fazlar\u0131n\u0131 profilleyerek CPU-bound g\u00f6revlerin nerede oldu\u011funu tespit etmek \u00f6nemlidir.<\/p>\n<h4>Soru 3: Worker Threads kullan\u0131rken dikkat edilmesi gereken en \u00f6nemli \u015fey nedir?<\/h4>\n<p><strong>Cevap:<\/strong> Worker Threads kullan\u0131rken dikkat edilmesi gereken en \u00f6nemli \u015fey, <strong>veri transferi maliyetini y\u00f6netmektir<\/strong>. \u0130\u015f par\u00e7ac\u0131klar\u0131 aras\u0131nda b\u00fcy\u00fck veri k\u00fcmelerini kopyalamak, Worker Threads'in sa\u011flad\u0131\u011f\u0131 performans kazanc\u0131n\u0131 kolayca yok edebilir. Bu nedenle, veriyi kopyalamak yerine aktar\u0131labilir nesneler (Transferable Objects) kullanarak sahipli\u011fi devretmek veya <code>SharedArrayBuffer<\/code> ile payla\u015f\u0131ml\u0131 bellek kullanarak veri transferi maliyetini minimize etmek kritik \u00f6neme sahiptir. Ayr\u0131ca, hata y\u00f6netimi ve worker'lar\u0131n ya\u015fam d\u00f6ng\u00fcs\u00fc y\u00f6netimini de g\u00f6z ard\u0131 etmemek gerekir.<\/p>\n<h4>Soru 4: <code class=\"language-\">require<\/code> veya <code class=\"language-\">import<\/code> ile worker dosyalar\u0131n\u0131 nas\u0131l y\u00f6netmeliyim?<\/h4>\n<p><strong>Cevap:<\/strong> Worker dosyalar\u0131, ana i\u015f par\u00e7ac\u0131\u011f\u0131ndan ba\u011f\u0131ms\u0131z bir ortamda \u00e7al\u0131\u015f\u0131r. Dolay\u0131s\u0131yla, worker dosyan\u0131z i\u00e7inde ihtiya\u00e7 duydu\u011funuz t\u00fcm mod\u00fclleri (<code>crypto<\/code>, <code>fs<\/code> vb.) do\u011frudan o worker dosyas\u0131 i\u00e7inde <code>require()<\/code> veya <code>import<\/code> etmelisiniz. Ana i\u015f par\u00e7ac\u0131\u011f\u0131nda y\u00fcklenen mod\u00fcller, otomatik olarak worker'lara aktar\u0131lmaz. Worker dosyas\u0131, sanki kendi ba\u015f\u0131na \u00e7al\u0131\u015fan ayr\u0131 bir Node.js beti\u011fi gibi d\u00fc\u015f\u00fcn\u00fclmelidir. Worker'\u0131 ba\u015flat\u0131rken <code>new Worker('.\/path\/to\/worker.js')<\/code> \u015feklinde do\u011fru dosya yolunu belirtti\u011finizden emin olun.<\/p>\n<h4>Soru 5: Worker Thread'ler taray\u0131c\u0131da \u00e7al\u0131\u015f\u0131r m\u0131?<\/h4>\n<p><strong>Cevap:<\/strong> Hay\u0131r, Node.js'in <code>worker_threads<\/code> mod\u00fcl\u00fc taray\u0131c\u0131 ortam\u0131nda \u00e7al\u0131\u015fmaz. Taray\u0131c\u0131lar, kendi \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 mekanizmalar\u0131na sahiptir ve bu mekanizmalar <strong>Web Workers<\/strong> olarak bilinir. Web Workers, Node.js Worker Threads'e benzer bir amaca hizmet eder (ana UI i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 bloklamadan arka planda CPU yo\u011fun g\u00f6revleri \u00e7al\u0131\u015ft\u0131rmak) ancak farkl\u0131 bir API ve g\u00fcvenlik modeli kullan\u0131r. Node.js Worker Threads, yaln\u0131zca Node.js \u00e7al\u0131\u015fma zaman\u0131 ortam\u0131 i\u00e7in tasarlanm\u0131\u015ft\u0131r.<\/p>\n<p><\/body><\/p>\n","protected":false},"excerpt":{"rendered":"Node.js’te \u00c7oklu \u0130\u015f Par\u00e7ac\u0131\u011f\u0131 Kullan\u0131m\u0131: Kapsaml\u0131 Rehber Node.js’in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz?…","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"csco_page_header_type":"","csco_page_load_nextpost":"","csco_page_subscribe_form":"","csco_page_contact_form":"","footnotes":""},"categories":[136],"tags":[],"class_list":{"0":"post-34630","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-node-js","7":"cs-entry","8":"cs-video-wrap"},"yoast_head":"\n<title>Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?<\/title>\n<meta name=\"description\" content=\"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?\" \/>\n<meta property=\"og:description\" content=\"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\" \/>\n<meta property=\"og:site_name\" content=\"Kodlar\u0131n Gizemli D\u00fcnyas\u0131\" \/>\n<meta property=\"article:published_time\" content=\"2025-11-19T21:21:56+00:00\" \/>\n<meta name=\"author\" content=\"Fatih Soysal\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Fatih Soysal\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"31 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\"},\"author\":{\"name\":\"Fatih Soysal\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"headline\":\"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?\",\"datePublished\":\"2025-11-19T21:21:56+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\"},\"wordCount\":5108,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"articleSection\":[\"Node.js\"],\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#respond\"]}],\"copyrightYear\":\"2025\",\"copyrightHolder\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\",\"name\":\"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\"},\"datePublished\":\"2025-11-19T21:21:56+00:00\",\"description\":\"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!\",\"breadcrumb\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Anasayfa\",\"item\":\"https:\/\/fatihsoysal.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/\",\"name\":\"Fatihsoysal.com\",\"description\":\"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim\",\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"tr\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\",\"name\":\"Fatih Soysal\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"contentUrl\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"width\":512,\"height\":512,\"caption\":\"Fatih Soysal\"},\"logo\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\"},\"description\":\"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/\"}]}<\/script>\n","yoast_head_json":{"title":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?","description":"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/","og_locale":"tr_TR","og_type":"article","og_title":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?","og_description":"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!","og_url":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/","og_site_name":"Kodlar\u0131n Gizemli D\u00fcnyas\u0131","article_published_time":"2025-11-19T21:21:56+00:00","author":"Fatih Soysal","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Fatih Soysal","Tahmini okuma s\u00fcresi":"31 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#article","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/"},"author":{"name":"Fatih Soysal","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"headline":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?","datePublished":"2025-11-19T21:21:56+00:00","mainEntityOfPage":{"@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/"},"wordCount":5108,"commentCount":0,"publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"articleSection":["Node.js"],"inLanguage":"tr","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#respond"]}],"copyrightYear":"2025","copyrightHolder":{"@id":"https:\/\/fatihsoysal.com\/blog\/#organization"}},{"@type":"WebPage","@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/","url":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/","name":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/#website"},"datePublished":"2025-11-19T21:21:56+00:00","description":"Node.js'in tek i\u015f par\u00e7ac\u0131kl\u0131 yap\u0131s\u0131n\u0131n getirdi\u011fi performans s\u0131n\u0131rlar\u0131n\u0131 a\u015fmak m\u0131 istiyorsunuz? Bu kapsaml\u0131 rehberde, Node.js'te \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) kullan\u0131m\u0131n\u0131, Worker Threads mod\u00fcl\u00fcn\u00fc ve ger\u00e7ek d\u00fcnya senaryolar\u0131n\u0131 ad\u0131m ad\u0131m ke\u015ffedin. Uygulamal\u0131 \u00f6rnekler ve ipu\u00e7lar\u0131yla uygulaman\u0131z\u0131n performans\u0131n\u0131 zirveye ta\u015f\u0131y\u0131n!","breadcrumb":{"@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Anasayfa","item":"https:\/\/fatihsoysal.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?"}]},{"@type":"WebSite","@id":"https:\/\/fatihsoysal.com\/blog\/#website","url":"https:\/\/fatihsoysal.com\/blog\/","name":"Fatihsoysal.com","description":"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim","publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"tr"},{"@type":["Person","Organization"],"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1","name":"Fatih Soysal","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","contentUrl":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","width":512,"height":512,"caption":"Fatih Soysal"},"logo":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/"},"description":"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.","url":"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/"}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/34630","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/comments?post=34630"}],"version-history":[{"count":0,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/34630\/revisions"}],"wp:attachment":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/media?parent=34630"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/categories?post=34630"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/tags?post=34630"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}

{"id":34630,"date":"2025-11-20T00:21:56","date_gmt":"2025-11-19T21:21:56","guid":{"rendered":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/"},"modified":"2025-11-20T00:21:56","modified_gmt":"2025-11-19T21:21:56","slug":"node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor","status":"publish","type":"post","link":"https:\/\/fatihsoysal.com\/blog\/node-js-neden-tek-is-parcacikli-ve-bu-ne-anlama-geliyor\/","title":{"rendered":"Node.js Neden Tek \u0130\u015f Par\u00e7ac\u0131kl\u0131 ve Bu Ne Anlama Geliyor?"},"content":{"rendered":"