operator design • Pipeline / Stage/ Task / Task Template / Version Control • Logging, monitoring, autoscaling, high availability • Extensibility / Integration • CI/CD examples • Future plan Our DevOps Pipeline/Stage/Task/Task Template/Version Control/UI generation/Volume... • Logging, monitoring, autoscaling, high availability • Extensibility/Integration • CI/CD examples • Future plan Overall Architecture created from template Tasks in same stage can run sequentially or in parallel Logging, Monitoring, Autoscaling, High Availability Kubernetes Cluster Node Node Node Node Job Job Job Job Pod Pod Pod ElasticSearch

✔ Load Balancing ✔ Overlay Networking ✔ Network Security Policies ✔ Backup and Recovery ✔ Autoscaling ✔ Service Discovery ✔ Networking ✔ RBAC & Access Control DEV DATA CENTER CLOUD BRANCH 5G / ✔ Load Balancing ✔ Overlay Networking ✔ Network Security Policies ✔ Backup and Recovery ✔ Autoscaling ✔ Service Discovery ✔ Networking ✔ RBAC & Access Control ✔ Common API & Packaging ✔ Health ✔ Load Balancing ✔ Overlay Networking ✔ Network Security Policies ✔ Backup and Recovery ✔ Autoscaling ✔ Service Discovery ✔ Networking ✔ RBAC & Access Control Common compute platform across any

分批策略 访问控制 流量配置 Deployment Function - componentName: frontend traits: - trait: apiVersion: autoscaling/v2beta2 kind: HorizontalPodAutoscaler spec: minReplicas: 1 maxReplicas: 10 - trait: apiVersion: spec: components: # 1st component - componentName: frontend traits: - trait: apiVersion: autoscaling/v2beta2 kind: HorizontalPodAutoscaler spec: minReplicas: 1 maxReplicas: 10 - trait: apiVersion:

allows for fine-grained autoscaling (including to/from zero) for event-driven Kubernetes workloads. KEDA serves as a Kubernetes Metrics Server and allows users to define autoscaling rules using a dedicated

specifications ● Kubernetes shortcomings with sidecar containers ○ Controlling containers lifecycle ○ Autoscaling pods with sidecar containers ● Are you prepared to handle Istio? ● A full mesh is utopian, know sidecar pattern in a better way, these workarounds should be deprecated. 21 Shortcoming 2: Autoscaling multi-containers pods Stabilizing Istio Kubernetes offers 2 ways to autoscale pods: ● HorizontalPodAutoscaler

automatically reload source code as it changes, including inotify(7) support on Linux. Read more... . • Autoscaling Dynamically resizing the worker pool depending on load, or custom metrics specified by the user • Remote control • Commands • Time Limits • Rate Limits • Max tasks per child setting • Autoscaling 92 Chapter 2. Contents Celery Documentation, Release 3.1.25 • Queues • Autoreloading • Inspecting using the workers –maxtasksperchild argument or using the CELERYD_MAX_TASKS_PER_CHILD setting. Autoscaling New in version 2.2. pool support: prefork, gevent The autoscaler component is used to dynamically

automatically reload source code as it changes, including inotify(7) support on Linux. Read more…. Autoscaling Dynamically resizing the worker pool depending on load, or custom metrics specified by the user paths Concurrency Remote control Commands Time Limits Rate Limits Max tasks per child setting Autoscaling Queues Autoreloading Inspecting workers Additional Commands Writing your own remote control commands using the workers –maxtasksperchild argument or using the CELERYD_MAX_TASKS_PER_CHILD setting. Autoscaling New in version 2.2. pool support: prefork, gevent The autoscaler component is used to dynamically

replication factor can be set to 1. in which case Kubernetes will bring it back up if it goes down. Autoscaling of pods replicas can be setup based on other conditions, such as CPU utilization. 1.3.6 Service almost all resource types. Auto scaling in Kubernetes is achieved by using the Horizontal Pod Autoscaling (HPA). Let’s take a closer look at how HPA works. • HPA Prerequisites: o HPA is triggered

• Time Limits • Rate Limits • Max tasks per child setting • Max memory per child setting • Autoscaling • Queues • Inspecting workers • Additional Commands • Writing your own remote control commands the workers --max-memory-per-child argument or using the worker_max_memory_per_child setting. Autoscaling New in version 2.2. pool support prefork, gevent The autoscaler component is used to dynamically needs two numbers: the maximum and minimum number of pool processes: --autoscale=AUTOSCALE Enable autoscaling by providing max_concurrency,min_concurrency. Example: --autoscale=10,3 (always keep 3 processes

control Commands Time Limits Rate Limits Max tasks per child setting Max memory per child setting Autoscaling Queues Inspecting workers Additional Commands Writing your own remote control commands Starting the workers --max-memory-per-child argument or using the worker_max_memory_per_child setting. Autoscaling New in version 2.2. pool support: prefork, gevent The autoscaler component is used to dynamically numbers: the maximum and minimum number of pool processes: --autoscale=AUTOSCALE Enable autoscaling by providing max_concurrency,min_concurrency. Example: --autoscale=10,3 (always keep