2025-02¶
- GitHub - Cyclenerd/google-cloud-compute-machine-types: ☁️ Choose the optimal Google Compute Engine machine type or instance in the many Google Cloud Platform regions
- GCE Machine Types in Google Cloud Platform
- GitHub - Privatehive/gcp-hosted-github-runner: ☁️ Automatically create/delete Google Cloud hosted ephemeral runner for GitHub Workflows
- Introduction to Machine Learning Interviews Book · MLIB
- GitHub - one2nc/cloudlens: k9s like CLI for AWS and GCP
- Create a virtual environment from a PEP 723 script · Issue #6637 · astral-sh/uv · GitHub PEP 723 metadata から仮想環境構築の議論
- GitHub - mrmathematica/venv-pack2: Package virtual environments for redistribution
- GitHub - flux-subsystem-argo/flamingo: Flux Subsystem for Argo
- GitOps Terraform Resources with Argo CD and Flux Subsystem for Argo - Flux Subsystem for Argo
- Customize EnvoyProxy | Envoy Gateway Envoy Gateway HPA
- GitHub - scality/runner-manager: Service to manage github Actions self-hosted runners デプロイメ周りのドキュメントはなさそう
- GitHub - jimmygchen/runner-fallback-action: Github action to determine the availability of self-hosted runners
- GitHub - vbem/multi-runners: Multi self-hosted runners on single host!
- GKE の ComputeClass に関する調査 | sreake.com | 株式会社スリーシェイク インスタンスが利用できない時に fallback の設定ができる
- rollouts-plugin-trafficrouter-gatewayapi/examples/envoygateway Argo Rollouts で Envoy Gateway を使う例
- Graduated and Incubating Projects | CNCF
- Sandbox Projects | CNCF
- GitHub - openkruise/kruise: Automated management of large-scale applications on Kubernetes (incubating project under CNCF)
- GitHub - karmada-io/karmada: Open, Multi-Cloud, Multi-Cluster Kubernetes Orchestration
-
Multi-Dimension HA: Region/AZ/Cluster/Provider
-
- GitHub - dragonflyoss/dragonfly: Dragonfly is an open source P2P-based file distribution and image acceleration system. It is hosted by the Cloud Native Computing Foundation (CNCF) as an Incubating Level Project.
- GitHub - cloud-custodian/cloud-custodian: Rules engine for cloud security, cost optimization, and governance, DSL in yaml for policies to query, filter, and take actions on resources
- GitHub - meshery/meshery: Meshery, the cloud native manager
- GitHub - kubereboot/kured: Kubernetes Reboot Daemon
- GitHub - koordinator-sh/koordinator: A QoS-based scheduling system brings optimal layout and status to workloads such as microservices, web services, big data jobs, AI jobs, etc.
- GitHub - Project-HAMi/HAMi: Heterogeneous AI Computing Virtualization Middleware
- Automated Dependency Updates for Terraform - Renovate Docs Terraform 内で docker provider を利用していれば、Renovate で更新できる
- GitHub - fresh2dev/TUIview: A TUI for every CLI. CLI を Textual TUI で動かす
- GitHub - helm/chart-testing: CLI tool for linting and testing Helm charts
- GitHub Copilot 活用術!#terminalLastCommand でコンテクストを補強しよう - Alternative Architecture DOJO
-
$ git diff -U9999 のあと、Chat で 「#terminalLastCommand コミットメッセージを作って」
-
- GitHub - pyper-dev/pyper: Concurrent Python made simple
- pyre-check/pyre2 at main · facebook/pyre-check · GitHub Python の型チェッカー pyre の v2。Rust で書き直されている
- HTTP Routing | Envoy Gateway
- Prerequisites で apply した quickstart のリソースを消さないとその先でエラーが出る。次に apply した時に
kubectl get gateways
を実行するとアドレスが付与されない - Envoy Gateway 経由でクラスタ内の svc にアクセスするには
curl http://<service-name>.<namespace>.svc.cluster.local
でアクセスできる
- Prerequisites で apply した quickstart のリソースを消さないとその先でエラーが出る。次に apply した時に
- GitHub - nicholasjackson/fake-service: Simple service for testing upstream service communications
shell # gRPCのfault injection $ docker run -p 8090:8090 -e LISTEN_ADDR=0.0.0.0:8090 -e ERROR_RATE=0.5 -e ERROR_TYPE=http_error -e ERROR_CODE=13 -e SERVER_TYPE=grpc nicholasjackson/fake-service:v0.26.2
- https://pypi.org/project/dataclass-alias/ pydantic のような dataclass の alias を提供するライブラリ。3rd party package に依存していないのが良い
- Envoy Gateway で Fault Injection を設定した上で retry が効いているか確認したところ、retry が効いておらず Fault Injection が優先されたような挙動になった
sequenceDiagram
participant C as Client
participant G as Gateway
participant S as Service
C->>G: Request
G->>S: Request
S->>G: Response 500
G->>S: Request
S->>G: Response 500
G->>S: Request
S->>G: Response 500
G->>C: Response 500
sequenceDiagram
participant C as Client
participant G as Gateway
participant S as Service
C->>G: Request
G->>C: Response 500
- akamai, linode のコンピューティングサービス。Spot Instance のようなものはないが、shared のプランが安価
- GitHub - arttor/helmify: Creates Helm chart from Kubernetes yaml
- Recommended way to add custom build steps · pypa/setuptools · Discussion #3762 · GitHub
- colima のデフォルトのリソースが少ない。割り当てるリソースを増やすには
colima start --cpu 4 --memory 8
のようにする。30 分かかっても終わらなかったdocker build
は、colima のリソースを増やしたらすぐに終わった - GitHub - secustor/backstage-plugins: contains backstage plugins written and maintained by myself
- GitHub - deepseek-ai/3FS: A high-performance distributed file system designed to address the challenges of AI training and inference workloads.
- GitHub - offerrall/photoff: GPU で画像処理を行う
- GitHub - deepseek-ai/smallpond: A lightweight data processing framework built on DuckDB and 3FS.
- GitHub - evidence-dev/evidence: Business intelligence as code: build fast, interactive data visualizations in SQL and markdown
- SWE bench を覗いてみたが、Python を使ってるレポジトリが多いように感じた。この指標はあらゆる言語に対して適用できるものではないのかも