ANALISIS PERBANDINGAN PERFORMA LATENCY DAN BIAYA LAYANAN SERVERLESS AWS LAMBDA DAN GOOGLE CLOUD FUNCTIONS DALAM PEMROSESAN DATA REAL-TIME: STUDI KASUS APLIKASI IOT

    Delia Fitri Setiani, - and Galura Muhammad Suranegara, - and Hafiyyan Putra Pratama, - (2025) ANALISIS PERBANDINGAN PERFORMA LATENCY DAN BIAYA LAYANAN SERVERLESS AWS LAMBDA DAN GOOGLE CLOUD FUNCTIONS DALAM PEMROSESAN DATA REAL-TIME: STUDI KASUS APLIKASI IOT. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Penelitian ini melakukan analisis perbandingan performa dari latency dan biaya AWS Lambda dan Google Cloud Functions (GCF) pada pemrosesan data real-time Internet of Things (IoT) menggunakan ESP32-CAM. Perangkat IoT mengambil gambar setiap 5.000 ms selama 7 hari secara berturut-turut, lalu dikirim kepada layanan serverless cloud pada AWS dan GCP melalui AWS API Gateway dan HTTP Trigger. Data gambar kemudian disimpan di Amazon S3 dan Google Cloud Storage (GCS). Data yang diambil untuk menganalisis latency adalah data timestamp pada proses pengiriman ke serverless, proses sampai di serverless dan proses tersimpan di cloud storage. Latency dianalisis secara end-to-end yaitu sebagai selisih waktu dari pengambilan gambar di perangkat IoT hingga objek tersimpan di storage (T3–T1). Analisis data menggunakan statistik deskriptif dan untuk dapat dibandingkan dan penghitungan biaya per 1.000 objek berdasarkan tagihan layanan. Hasil menunjukkan volume pengambilan data yang diperoses pada AWS dan GCP cukup sebanding untuk beban ringan. Rata-rata latency end-to-end AWS lebih rendah dibanding GCP, dengan bottleneck utama berasal dari perangkat atau jaringan, bukan waktu eksekusi serverless. Biaya pada beban ringan didominasi oleh operasi dan retensi penyimpanan objek. _____ This study presents a comparative analysis of latency and cost between AWS Lambda and Google Cloud Functions (GCF) for real-time Internet of Things (IoT) processing using an ESP32-CAM. The device captured and transmitted an image every 5,000 ms (5 s) for seven consecutive days to both serverless services via AWS API Gateway and the GCF HTTP trigger. The resulting objects were stored in Amazon S3 and Google Cloud Storage (GCS). End-to-end latency was defined as the elapsed time from image capture on the device until the object is durably stored (T3–T1), using timestamps recorded at dispatch to the serverless endpoint, at arrival/start within the function, and at write completion in cloud storage. Data were analyzed using descriptive statistics, and costs were computed per 1,000 objects from provider billing records to enable direct comparison. The results show that, under a light workload, both platforms processed a comparable volume of data. However, average end-to-end latency on AWS was lower than on GCF in this scenario, with the dominant bottlenecks attributable to the device or network paths rather than function execution time. Under light load, total cost is driven primarily by object-storage operations and retention rather than compute.

    Baca Full Text klik disini
    [thumbnail of S_SISTEL_2104506_Title.pdf] Text
    S_SISTEL_2104506_Title.pdf
    Download (806kB)
    [thumbnail of S_SISTEL_2104506_Chapter1.pdf] Text
    S_SISTEL_2104506_Chapter1.pdf
    Download (430kB)
    [thumbnail of S_SISTEL_2104506_Chapter2.pdf] Text
    S_SISTEL_2104506_Chapter2.pdf
    Restricted to Staf Perpustakaan
    Download (506kB)
    [thumbnail of S_SISTEL_2104506_Chapter3.pdf] Text
    S_SISTEL_2104506_Chapter3.pdf
    Download (571kB)
    [thumbnail of S_SISTEL_2104506_Chapter4.pdf] Text
    S_SISTEL_2104506_Chapter4.pdf
    Restricted to Staf Perpustakaan
    Download (775kB)
    [thumbnail of S_SISTEL_2104506_Chapter5.pdf] Text
    S_SISTEL_2104506_Chapter5.pdf
    Download (535kB)
    [thumbnail of S_SISTEL_2104506_Appendix.pdf] Text
    S_SISTEL_2104506_Appendix.pdf
    Restricted to Staf Perpustakaan
    Download (977kB)
    Official URL: https://repository.upi.edu/
    Item Type: Thesis (S1)
    Additional Information: https://scholar.google.com/citations?user=zJSs578AAAAJ&hl=en ID SINTA Dosen Pembimbing: Galura Muhammad Suranegara: 6703764 Hafiyyan Putra Pratama: 6681148
    Uncontrolled Keywords: Serverless, AWS Lambda, Google Cloud Functions, IoT, latency, cost(Biaya).
    Subjects: L Education > L Education (General)
    T Technology > T Technology (General)
    Divisions: UPI Kampus Purwakarta > S1 Sistem Telekomunikasi
    Depositing User: Delia Fitri Setiani
    Date Deposited: 01 Sep 2025 04:44
    Last Modified: 01 Sep 2025 04:44
    URI: http://repository.upi.edu/id/eprint/136981

    Actions (login required)

    View Item View Item