(Vasia) Kalavri
 vkalavri@bu.edu Spring 2020 4/28: Graph Streaming ??? Vasiliki Kalavri | Boston University 2020 Modeling the world as a graph 2 Social networks friend follows The web Actor-movie results for the search term “graph” ??? Vasiliki Kalavri | Boston University 2020 Basics 1 5 4 3 2 “node” or “vertex” “edge” 1 5 4 3 2 undirected graph directed graph 4 ??? Vasiliki Kalavri Kalavri | Boston University 2020 Graph streams Graph streams model interactions as events that update an underlying graph structure 5 Edge events: A purchase, a movie rating, a like on an online post

https://mp.weixin.qq.com/mp/appmsgalbum?biz=MzUzNzYxNjAzMg==&action =getalbum&album_id=1416643008195608577&scene=173&from_msgid=2247 491488&from_itemidx=1&count=3#wechat_redirect OAM 社区会议记录： https://space.bilibili

古思为 ⽅阗 Graph DB on K8s Demystified and its Serverless applicaiton in actions. DEVELOPER ADVOCATE @ MAINTAINER OF KCD China 2021 Nov. 6th @Shanghai 古思为 wey-gu wey_gu siwei.io Nebula Graph 开发者布道师 示存 储数据的数据库 wikipedia.org/wiki/graph_database 了解更多关于 什么是图数据库 什么是图数据库 为什么需要图数据库？ 传统数据库 图数据库 图模型的结构 图语义的查询 性能 Nebula Graph! 如何发⾳：[ˈnebjələ]， 它有哪些特点？ Nebula Graph 介绍 了解更多 >>> ⽂档：Nebula 架构 官⽹：⽤户案例 ├── hack ├── doc └── tests └── e2e name: statefulsets.apps.kruise.io apiVersion: apps.nebula�graph.io/v1alpha1 kind: NebulaCluster metadata: name: nebula spec: graphd: resources: requests: cpu:

cincinnati-graph-data.tar.gz https://github.com/openshift/cincinnati-graph- data/archive/master.tar.gz CMD exec /bin/bash -c "tar xvzf cincinnati-graph-data.tar.gz -C /var/lib/cincinnati/graph-data/ - com/openshift/graph-data:latest： 3. 将上一步中创建的 graph-data 容器镜像推送到 OpenShift Update Service 可以访问的存储库， 如 registry.example.com/openshift/graph-data:latest： 注意 注意 要将图形数据镜像推送到受限网络中的本地 registry，请将上一步中创建的 graph- data $ podman build -f ./Dockerfile -t registry.example.com/openshift/graph-data:latest $ podman push registry.example.com/openshift/graph-data:latest x509: certificate relies on legacy Common Name field

务器的 URL 设置为 upstream，以便在更新期间使用本地服务器。upstream 的默认值是 https://api.openshift.com/api/upgrades_info/v1/graph。 流程 流程 更改集群版本中的 upstream 参数值： 变量指定更新服务器的 URL。 输出示例 出示例 3.2. 使用 WEB 控制台更新集群 变量指定更新服务器的 URL。 默认 upstream 是 https://api.openshift.com/api/upgrades_info/v1/graph。 3. 点击 Save。 其他 其他资 资源 源 了解更新频道和发行版本 3.2.3. 使用 Web 控制台暂停 MachineHealthCheck 资源 在更新过程中，集群中的节点可能会临时不可用。对于 将频道设置为从中检索 OpenShift Container Platform 镜像。 添加 graph: true 以构建并推送 graph-data 镜像推送到镜像 registry。创建 OpenShift Update Service (OSUS) 需要 graph-data 镜像。graph: true 字段还会生成 UpdateService 自定义资源清单。oc 命令行界面 (CLI)

(and advantages) • Preview of Numpy & PyTorch & Tensorflow Numpy Tensorflow PyTorch Computation Graph Advantages (continued) • Which one do you think is better? Advantages (continued) • Which one • Visualization Tools like • TensorboardX (monitor training) • PyTorchViz (visualise computation graph) • Various other functions • loss (MSE,CE etc..) • optimizers Prepare Input Data •Load data •Iterate to run on. Visualization • TensorboardX (visualise training) • PyTorchViz (visualise computation graph) https://github.com/lanpa/tensorboardX/ Visualization (continued) • PyTorchViz https://github

basics 3 source sink input port output port dataflow graph ??? Vasiliki Kalavri | Boston University 2020 Revisiting the basics 4 Dataflow graph • operators are nodes, data channels are edges • 0.5 Operator re-ordering B A A B ??? Vasiliki Kalavri | Boston University 2020 17 • A static graph transformation that enables re-ordering at runtime • It dynamically routes data after measuring Kalavri | Boston University 2020 22 • Multi-tenancy • in streaming systems that build one dataflow graph for several queries • when applications analyze data streams from a small set of sources • Operator

University 2020 src o1 o2 10 recs 10 recs 1 2 3 4 100 rec 100 recs Intuition: use the dataflow graph to extract operator dependencies and system instrumentation to collect accurate, representative University 2020 src o1 o2 10 recs 10 recs 1 2 3 4 100 rec 100 recs Intuition: use the dataflow graph to extract operator dependencies and system instrumentation to collect accurate, representative University 2020 src o1 o2 10 recs 10 recs 1 2 3 4 100 rec 100 recs Intuition: use the dataflow graph to extract operator dependencies and system instrumentation to collect accurate, representative

Kiali 图形的布局会根据应用程序构架和要显示的数据（图形节点数目及其交互）的不 同而有所变化。因为创建一个统一布局的难度较大，所以 Kiali 提供了几种不同布局的选择。要选 择不同的布局，可从 Graph Settings 菜单中选择一个不同的 Layout Schema。 首次从 Kiali 控制台访问相关服务（如分布式追踪平台和 Grafana）时，必须使用 OpenShift Container 控制台中，'istio-ingressgateway' 和 'jaeger-query' 服务的服务详情页面中没 有显示 Traces。Jaeger 中存在 trace。 OSSM-287 在 Kiali 控制台中没有显示 Graph 服务中的 trace。 OSSM-285 试图访问 Kiali 控制台时会收到以下错误消息："Error trying to get OAuth Metadata"。 临时解决方案：重启 Kiali 要查看 Service Mesh control plane 节点，点 Graph 页面，从菜单中选择安装 ServiceMeshControlPlane 的命名空间，如 istio-system。 a. 如有必要，请单击 Display idle nodes。 b. 要了解更多有关 Graph 页面的信息，请点击 Graph tour 链接。 c. 要查看网格拓扑，请从 Namespace

分布式存储  分布式查询 功能完备:  GSL/负采样  主流图算法  异构图 (user/item/attribute)  动态图 标准化: Standard Libraries Graph-Learn: 分布式图算法库 标准化: Standard Solutions Continuous Optimization: Active learning Data Label Model [VariationalDropout] 通信优化 [GRPC++] 实时训练 [增量更新] 混合精度 [bf16] 工程优化: 千亿特征优化 模型蒸馏 AVX/SSE优化 Graph优化 [User Graph去重] 内存Allocate优化 ParallelStringOp [split/type conversion] Sequence Feature [side info] com/alibaba/EasyCV 6. EasyNLP: https://github.com/alibaba/EasyNLP 7. AliGraph: https://github.com/alibaba/graph-learn 8. DSW: https://help.aliyun.com/document_detail/194831.html 9. DLC: https://help.aliyun.c