流式k-means 算法 1 流式k-means 算法原理 2 流式k-means 算法源码分析 流式k-means 算法   当数据是以流的方式到达的时候，我们可能想动态的估计（estimate ）聚类的簇，通过新的到达的数据来更新聚类。spark.mllib 支持流式k-means 聚类，并且可以通过参数控制估计衰减（decay ）(或“健忘...

Approximate k-NN search Recommendations for engines and cluster node sizing Get started with approximate k-NN Building a k-NN index from a model Using approximate k-NN with filt...

k-NN vector quantization Lucene byte vector Lucene scalar quantization Using Lucene scalar quantization Confidence interval Memory estimation HNSW memory estimation Faiss 16-...

k-NN search with filters Filtered search optimization Efficient k-NN filtering Lucene k-NN filter implementation Using a Lucene k-NN filter Faiss k-NN filter implementation Usi...

Approximate k-NN search Recommendations for engines and cluster node sizing Get started with approximate k-NN The number of returned results Building a k-NN index from a model A...

k-NN vector quantization Lucene byte vector Faiss 16-bit scalar quantization Using Faiss scalar quantization Memory estimation HNSW memory estimation IVF memory estimation Fa...

k-NN search with filters Filtered search optimization Efficient k-NN filtering Lucene k-NN filter implementation Using a Lucene k-NN filter Faiss k-NN filter implementation Usi...

k-NN vector quantization Lucene byte vector Faiss 16-bit scalar quantization Using Faiss scalar quantization Memory estimation HNSW memory estimation IVF memory estimation Fa...

监控K8s 前置依赖 通过kubelet来了解和监控k8s节点的基本运行状态数据 方案一：直接访问kubelet来获取节点状态指标数据 方案二：通过kube-apiserver提供的API间接获取kubelet的指标数据 通过cAdvisor来了解和监控节点中的容器运行状态 方案一：直接访问kubelet的/metrics/cadvisor地址，需...

k8s部署 一. 使用 h2 作为数据库 1. 创建 nameSpace 和 configMap 2. 部署 shenyu-admin 3. 部署 shenyu-bootstrap 二. 使用 mysql 作为数据库 1. 创建 nameSpace和 configMap 2. 创建 endpoint 代理外部 mysql 3. 创建 pv 存储 ...