Takizawa Takeshi from Rakuten Group discusses the development of cloud-native CI/CD services, emphasizing the use of technologies like containers and microservices to enhance scalability in cloud environments. The presentation outlines the current infrastructure, including Kubernetes and Jenkins setup, while highlighting challenges faced due to Jenkins' legacy status. Proposed solutions focus on dynamic provisioning of CI/CD tools to improve efficiency and address existing operational issues.
The document discusses recent advances in generative adversarial networks (GANs) for image generation. It summarizes two influential GAN models: ProgressiveGAN (Karras et al., 2018) and BigGAN (Brock et al., 2019). ProgressiveGAN introduced progressive growing of GANs to produce high resolution images. BigGAN scaled up GAN training through techniques like large batch sizes and regularization methods to generate high fidelity natural images. The document also discusses using GANs to generate full-body, high-resolution anime characters and adding motion through structure-conditional GANs.
作って(壊して?)学ふ?インターネットのしくみ サイハ?ーエーシ?ェントの実験用ASの紹介 / Introduce experimental AS in ...whywaita
?
サイバーエージェントの社内エンジニアカンファレンス CA BASE CAMP 2021で発表した資料です
再アップロード版: https://speakerdeck.com/whywaita/as-introduce-experimental-as-in-cyberagent-internet-seminer
23. Service
? Podの起動停止に合わせたサービスディスカバリ
? Pod宛てのトラフィックをロードバランシングする
? ロードバランシングの接続先仮想IPアドレスの付与
? クラスタ内DNSへの自動登録
? 外部ロードバランサーへの登録
22
Service
Pod
Pod
Pod
Pod
外部LoadBalancer
10.42.0.138
10.42.0.43
10.42.0.110
ClusterIP
10.43.10.132 分散される
新しく追加されるとメンバー
に自動的に追加される