DOCSEARCHbyAlgolia
Submit Docs

KakaoCloud Documentation

website·Indexed 33 days ago

https://docs.kakaocloud.com/

Cloud computing platform documentation featuring guides on infrastructure, big data pipelines, and AI/ML workflows. Includes tutorials for Kubeflow traffic prediction, CDC pipelines with Debezium and Kafka, Hadoop Eco clusters, OpenSearch-based AI use cases, and observability integrations with Splunk and Zabbix.

Tokens
1.3M
Snippets
4K
Records
9.8K
Agent score
60%

What's inside KakaoCloud

  1. Overview of Codeflow and Deployflow for CI/CD

    Kakao Cloud provides Codeflow and Deployflow to connect the development and deployment process into a single, traceable pipeline. This integration allows developers to track how a specific code commit leads to a specific container image and manifest, and finally to a deployment in a Kubernetes cluster. It is particularly useful for teams needing clear traceability, approval processes before deployment, and advanced deployment strategies like Blue/Green or Canary.

  2. Overview of Kakao Cloud Certificate Manager

    Certificate Manager is a centralized management service for SSL certificates within Kakao Cloud. It allows users to register certificates in a single console and link them to multiple services, eliminating the need to configure certificates individually for each service. This ensures security consistency and improves operational efficiency.

  3. Overview of Kakao Cloud Native CI/CD Pipeline Flow

    The native CI/CD pipeline using Codeflow and Deployflow follows these five stages:

    1. Source Management (Codeflow): Application source code and Kubernetes manifests are stored in a Codeflow repository.
    2. Image Build (Codeflow Actions): A Codeflow workflow triggers a self-hosted runner to build the container image.
    3. Image Upload (Container Registry): The built image is pushed to the Kakao Cloud Container Registry.
    4. Manifest Update (Codeflow Actions): The workflow updates the deployment manifest (k8s/deployment.yaml) with the new image tag and commits the change back to the repository.
    5. Deployment Automation (Deployflow): Deployflow detects the manifest change and synchronizes it with the Kubernetes Engine cluster, which then pulls the image from the Container Registry to run the Pods.

  4. Overview of r2a memory-optimized instances

    The r2a instance family is designed for memory-intensive workloads that process large-scale data in memory. These instances utilize 3rd generation AMD EPYC 7000 series processors (specifically AMD EPYC 7643) with an all-core turbo clock speed up to 3.6GHz. They support AMD instruction sets AVX and AVX2.

    Recommended Use Cases:

    • High-performance relational and NoSQL databases
    • Distributed in-memory caches (e.g., Redis, Memcached)
    • Real-time big data analytics (e.g., Hadoop, Spark clusters)
    • High-Performance Computing (HPC) and Electronic Design Automation (EDA) applications.

  5. Overview of Kakao Cloud Hadoop Eco

    Hadoop Eco is a cloud platform service for executing distributed processing tasks using open-source frameworks such as Hadoop, Hive, HBase, Spark, Trino, and Kafka. It provides provisioning services for Hadoop, HBase, Trino, and Dataflow using Kakao Cloud Virtual Machines (VMs).

  6. Overview of KakaoCloud Resource Explorer

    Resource Explorer is a management service in KakaoCloud that provides an integrated view of various cloud resources within a project. It allows users to identify, search, and manage assets such as instances, Block Storage, public IPs, load balancers, and backups from a single screen instead of navigating individual service consoles.

  7. Overview of Kakao Cloud Maintenance service

    The Maintenance service allows users to systematically manage OS, engine, and kernel upgrades, as well as configuration changes for instances. It provides a UI to create tasks, schedule resources, monitor execution status, and receive notifications to prevent unexpected failures and improve operational efficiency.

  8. Overview of KakaoCloud Role Inheritance

    Role inheritance allows administrators to assign project roles to users or groups across all projects within an organization. This eliminates the need to assign roles separately for each project, and any roles assigned via inheritance are automatically applied to projects created in the future. Management is performed via the KakaoCloud Console at Management > IAM > Roles > Role inheritance.

  9. Overview of KakaoCloud AI Insight

    AI Insight is an AI monitoring service that provides an integrated view of resource status and key metrics for AI workload operations across clusters, nodes, and GPUs. It helps operators identify whether resources are active, idle, or abnormal, which is critical for managing performance and costs in AI model training and inference environments.

  10. Overview of Kakao Cloud VM Import service

    The VM Import service allows users to migrate VM instances, custom images, or disks from existing environments into the Kakao Cloud Compute service. This enables efficient migration and management of existing virtualization images in the cloud. Note that the VM Import service is provided exclusively within private network environments.

  11. Overview of Codeflow and Deployflow for CI/CD

    Codeflow and Deployflow are Kakao Cloud services designed to create a traceable and reviewable CI/CD pipeline. Together, they connect source code management, automated builds, image storage, and Kubernetes deployment within the Kakao Cloud ecosystem. This integration allows developers to track the entire flow from a specific code commit to the resulting container image, the updated manifest, and the final deployment state in a cluster.

  12. Overview of Advanced Managed Prometheus

    Advanced Managed Prometheus is a high-performance managed monitoring service by KakaoCloud designed for cloud-native environments. It collects, stores, and analyzes metric data from Kubernetes, Virtual Machines (VMs), and applications. The service automates the installation, configuration, and backup of Prometheus to reduce operational overhead and ensure scalability and stability for large-scale data processing.