2 个仓库
Tools for ensuring consistent system-level configurations like hostnames, timezones, and networking across all nodes.
Distinct from Definition and Standardization: Existing candidates focused on container standards or blueprint definitions rather than the actual standardization of host-level OS settings.
Explore 2 awesome GitHub repositories matching devops & infrastructure · Node Environment Standardization. Refine with filters or upvote what's useful.
Pigsty 是一个全面的数据库基础设施编排平台,旨在自动化高可用 PostgreSQL 集群的全生命周期。它作为一个基础设施即代码(IaC)框架,通过幂等 Playbook 管理集群协调、节点配置与服务发现。通过集成分布式共识机制,该平台确保了在包括裸机与虚拟化基础设施在内的多样化环境中,自动化故障转移与一致的状态强制执行。 该平台通过一套超越标准数据库管理的强大运营能力脱颖而出。它具备内置的观测流水线,将指标、日志与追踪聚合到集中式仪表盘中,用于实时性能监控与诊断分析。此外,它还提供了一个模拟专有线路协议与 SQL 语法的迁移框架,允许将遗留企业数据库工作负载集成到现代关系型环境中。 该系统涵盖了广泛的功能面,包括带有写时复制(CoW)克隆以实现快速部署的高级存储管理,以及协调关系型引擎与缓存及对象存储服务的多数据库编排。它还整合了安全加固、自动化备份与恢复,以及通过分层代理进行的流量路由,以将客户端连接与底层集群拓扑解耦。 该项目以自包含的包镜像模型分发,能够在安全或离线环境中实现一致的部署与依赖管理。
Converges system-level settings for hostnames, timezones, and core utilities to ensure consistency across nodes.
Higgsfield is a distributed machine learning framework designed to scale the training of neural networks with billions of parameters across large-scale GPU clusters. It provides the infrastructure necessary to orchestrate complex computational workflows, manage heterogeneous compute resources, and automate the deployment of training tasks across multiple nodes. The platform distinguishes itself through advanced distributed training strategies, including parameter sharding to accommodate models that exceed individual hardware memory capacities and asynchronous gradient aggregation to optimize
Standardizes dependencies and configurations across multiple compute nodes to ensure consistent execution and reliable results for data processing tasks.