diff --git a/.github/workflows/tca-release.yml b/.github/workflows/tca-release.yml index 373b693..8d853ba 100644 --- a/.github/workflows/tca-release.yml +++ b/.github/workflows/tca-release.yml @@ -4,6 +4,10 @@ on: push: tags: - "v*.*.*" + +permissions: + contents: write + env: IMAGE_REGISTRY: quay.io @@ -66,4 +70,4 @@ jobs: - name : tag image run : docker tag tackle:latest quay.io/konveyor/tackle-container-advisor:${{ github.ref_name }} - name : push to quay - run : docker push quay.io/konveyor/tackle-container-advisor:${{ github.ref_name }} \ No newline at end of file + run : docker push quay.io/konveyor/tackle-container-advisor:${{ github.ref_name }} diff --git a/README.md b/README.md index 98dcfab..697c9ef 100644 --- a/README.md +++ b/README.md @@ -3,51 +3,58 @@ ### Usage 1. [TCA Pipeline](#TCA-Pipeline) -2. [Setup and Running TCA](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/running.md) -3. [API's](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/service.md) -4. [References](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/references.md) - - +2. [Run TCA](https://konveyor.github.io/tackle-container-advisor-docs/docs/setup/) +3. [API description](https://konveyor.github.io/tackle-container-advisor-docs/docs/apis) +4. [References](https://konveyor.github.io/tackle-container-advisor-docs/docs/publications/) ### Development -1. [Knowledge Base](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/db.md) -2. [KG Utils](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/utils.md) -3. [Entity standardizer](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/standardizer.md) - +1. [Knowledge Base](https://konveyor.github.io/tackle-container-advisor-docs/docs/kg/) +2. [Knowledge Base Utilities](https://konveyor.github.io/tackle-container-advisor-docs/docs/kgutils/) +3. [Entity standardizer](https://konveyor.github.io/tackle-container-advisor-docs/docs/standardizer/) +4. [Developer Guide](https://konveyor.github.io/tackle-container-advisor-docs/docs/development/) ## Purpose - -TCA takes client applications as a natural language description and recommends whether client applications can be containerized. For example, a client can provide the application description as the following. - +TCA provides APIs to standardize natural language description of technology stack components, cluster a portfolio of technology stacks into similar technology stack groups, match technology stacks to docker, openshift or operator catalog images. ``` 1. App1: rhel, db2, java, tomcat +2. App2: .net, java, oracle db +3. App3: dot net, java, oracle dbms ``` - TCA takes the following steps to recommend the containerization. -1. **Standardization**: It assesses the application to standardize the inputs to relevant named entities present in our knowledge base. For details on the knowledge base please check the [Knowledge Base](https://github.com/konveyor/tackle-container-advisor/blob/main/docs/docs/db.md). -For example, the inputs in *App1* get mapped as the following named entities. +**Standardize**: Standardize natural language inputs to relevant named entities of technology stacks present in our knowledge base. For details on the knowledge base please check the *db* folder. For example, the inputs in *App1,App2,App3* get mapped as the following named entities. ``` -1. App1: rhel: Linux|RedHat Linux, db2: DB2, java: Java, tomcat: Apache Tomcat +1. App1: rhel: RedHat Enterprise Linux, db2: DB2, java: Java, tomcat: Apache Tomcat +2. App2: .net: .NET, java: Java, oracle db: Oracle DB +3. App3: dot net: .NET, java: Java, oracle dbms: Oracle DB ``` -2. **Containerization**: First, it recommends whether *App1* can be containerized, partially containerized, or kept as it is. Then if App1 is recommended as containerized or partially containerized, TCA generates container images based on DockerHub or Openshift. For example, if a user decides to generate DockerHub related images, then TCA generates the following images. +**Clustering**: Cluster the standardized technology stack components into groups of similar technology stacks. For example, the standardized technology stacks for *App1,App2,App3* get clustered into the two technology stack clusters. ``` -1. tomcat|https://hub.docker.com/_/tomcat -2. db2|https://hub.docker.com/r/ibmcom/db2 +1. Cluster1: {App1} +2. Cluster2: {App2, App3} +``` + +**Containerize**: Determines whether a technology stack is fully containerizable, partially containerizable or not containerizableat all. If a technology stack is recommended as fully or partially containerizable, it also generates container images based on DockerHub or Openshift image catalogs. It is also possible to provide custom user-defined catalogs for matching to catalog images. For example, if a user decides to generate DockerHub related images, then TCA generates the following images. + +``` +1. Cluster1: tomcat|https://hub.docker.com/_/tomcats +2. Cluster2: db2|https://hub.docker.com/r/ibmcom/db2 ``` For OpenShift, TCA generates the following images. - 1. tomcat|https://access.redhat.com/containers/#/registry.access.redhat.com/jboss-webserver-3/webserver31-tomcat8-openshift - 2. db2|https://access.redhat.com/containers/#/cp.stg.icr.io/cp/ftm/base/ftm-db2-base +``` +1. Cluster1: tomcat|https://access.redhat.com/containers/#/registry.access.redhat.com/jboss-webserver-3/webserver31-tomcat8-openshift +2. Cluster2: db2|https://access.redhat.com/containers/#/cp.stg.icr.io/cp/ftm/base/ftm-db2-base +``` ## TCA Pipeline -TCA Pipeline +![TCA Pipeline](images/tca_pipeline.png) The pipeline ingests raw inputs from clients data and standardizes the data to generate named entities and versions. For standardizing or normalizing raw inputs we use a tf-idf similarity based approach. To find container images we represent images in terms of named entities as well. The normalized representation helps to match legacy applications with container images to suggest the best possible recommendations.