From 22845c936f0cfc69b59522c3356ead45eeae6ffc Mon Sep 17 00:00:00 2001 From: Jakub Scholz Date: Mon, 19 Mar 2018 10:16:56 +0100 Subject: [PATCH 1/3] Add Roadmap page --- _data/menu.yml | 3 +++ roadmap/index.md | 65 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 68 insertions(+) create mode 100644 roadmap/index.md diff --git a/_data/menu.yml b/_data/menu.yml index c0b1ee38e..459c7c597 100644 --- a/_data/menu.yml +++ b/_data/menu.yml @@ -2,6 +2,9 @@ header: - url: / title: Overview identifier: overview + - url: /roadmap + title: Roadmap + identifier: roadmap - url: /blog title: Blog identifier: blog \ No newline at end of file diff --git a/roadmap/index.md b/roadmap/index.md new file mode 100644 index 000000000..9342d9031 --- /dev/null +++ b/roadmap/index.md @@ -0,0 +1,65 @@ +--- +# You don't need to edit this file, it's empty on purpose. +# Edit theme's home layout instead if you wanna make some changes +# See: https://jekyllrb.com/docs/themes/#overriding-theme-defaults +layout: default +--- + +# Roadmap + +## Improve configuration possibilities for Kafka and Kafka Connect + +Currently, Strimzi gives the user only limited possibilities to configure Kafka and Kafka Connect. Only few +configuration options are exposed and user configurable. Users should have more freedom to fine tune Kafka and +Kafka Connect configuration according to their exact needs. + +## Add support for encryption, authentication and authorization + +In order to use Strimzi for production workloads, it has to be able to properly secure the cluster. This should +include: + +* Encryption using TLS +* Authentication using TLS client authentication and SASL +* Authorization + +## Accessing Kafka from the outside of Kubernetes / OpenShift + +Currently the Kafka deployment is accessible only from within the same Kubernetes / OpenShift cluster in which it is +deployed. However in some scenarios it might be useful to access it also from the outside. + +## Add support for Kubernetes resource request and limits + +All Strimzi deployments are currently running without any resource requests and limits (see +[Kubernetes documentation](https://kubernetes.io/docs/concepts/configuration/manage-compute-resources-container/) +for more details). A possibility to configure resource limits and requests should be added to the Cluster Controller +component. + +## Add support for Kubernetes node selectors + +In order to achieve good performnce, it might be needed to schedule Kafka pods to specific nodes. For example because +they contain special storage hardware or just to run them on dedicated nodes to make sure that KAfka doesn't interfere +with any other applications. To achiave this, it should be possible to configure node selectors for the Strimzi +deployments. + +## Custom Resource Definitions support + +Cluster and Topic Controllers should have support for [Custom Resource Definitions (CRDs)](https://kubernetes.io/docs/concepts/api-extension/custom-resources/). + +## Service broker suport + +Cluster Controller should be able to work as a [Service Broker](https://www.openservicebrokerapi.org/). + +## Support for automated cluster balancing + +During the lifecycle of the Kafka cluster it can happen that it becomes unbalanced. Some nodes are hosting very _heavy_ +topics (i.e. busy topics with a lot of traffic) whicle other nodes are idle most of the time hosting less busy topics. +Automated cluster balancer should continuously monitor the cluster state and balance it (re-distribute the topics) when +needed to make sure that the load is optimally distributed across all cluster nodes. + +## Kafka updates + +Strimzi should make it possible to smoothly handle updates from one Kafka version to another. + +## Integration with other protocols + +Allow to access Kafka using different protocols such as HTTP, AMQP or MQTT. \ No newline at end of file From 32527a3fda441e1c0af041d1e7eaf691d5fc6a53 Mon Sep 17 00:00:00 2001 From: Jakub Scholz Date: Mon, 19 Mar 2018 10:30:28 +0100 Subject: [PATCH 2/3] Fix typos --- roadmap/index.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/roadmap/index.md b/roadmap/index.md index 9342d9031..1b872d160 100644 --- a/roadmap/index.md +++ b/roadmap/index.md @@ -37,7 +37,7 @@ component. ## Add support for Kubernetes node selectors In order to achieve good performnce, it might be needed to schedule Kafka pods to specific nodes. For example because -they contain special storage hardware or just to run them on dedicated nodes to make sure that KAfka doesn't interfere +they contain special storage hardware or just to run them on dedicated nodes to make sure that Kafka doesn't interfere with any other applications. To achiave this, it should be possible to configure node selectors for the Strimzi deployments. @@ -45,7 +45,7 @@ deployments. Cluster and Topic Controllers should have support for [Custom Resource Definitions (CRDs)](https://kubernetes.io/docs/concepts/api-extension/custom-resources/). -## Service broker suport +## Service broker support Cluster Controller should be able to work as a [Service Broker](https://www.openservicebrokerapi.org/). From 1f86a3e2c0cc1612a1278ddac0b431dc5054bbd1 Mon Sep 17 00:00:00 2001 From: Tom Bentley Date: Mon, 19 Mar 2018 09:55:01 +0000 Subject: [PATCH 3/3] Minor changes to roadmap --- roadmap/index.md | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) diff --git a/roadmap/index.md b/roadmap/index.md index 1b872d160..ef5c5aa9d 100644 --- a/roadmap/index.md +++ b/roadmap/index.md @@ -1,7 +1,4 @@ --- -# You don't need to edit this file, it's empty on purpose. -# Edit theme's home layout instead if you wanna make some changes -# See: https://jekyllrb.com/docs/themes/#overriding-theme-defaults layout: default --- @@ -9,23 +6,23 @@ layout: default ## Improve configuration possibilities for Kafka and Kafka Connect -Currently, Strimzi gives the user only limited possibilities to configure Kafka and Kafka Connect. Only few -configuration options are exposed and user configurable. Users should have more freedom to fine tune Kafka and +Currently, Strimzi gives the user only limited possibilities to configure Kafka and Kafka Connect. Only a few +configuration options are exposed and user configurable. Users should have more freedom to fine-tune Kafka and Kafka Connect configuration according to their exact needs. ## Add support for encryption, authentication and authorization -In order to use Strimzi for production workloads, it has to be able to properly secure the cluster. This should +In order to use Strimzi for production workloads, it has to be possible to secure the cluster. This should include: * Encryption using TLS * Authentication using TLS client authentication and SASL * Authorization -## Accessing Kafka from the outside of Kubernetes / OpenShift +## Accessing Kafka from the outside of Kubernetes/OpenShift -Currently the Kafka deployment is accessible only from within the same Kubernetes / OpenShift cluster in which it is -deployed. However in some scenarios it might be useful to access it also from the outside. +Currently, the Kafka deployment is accessible only from within the same Kubernetes/OpenShift cluster in which it is +deployed. In some scenarios it will be necessary to access it from the outside. ## Add support for Kubernetes resource request and limits @@ -36,24 +33,24 @@ component. ## Add support for Kubernetes node selectors -In order to achieve good performnce, it might be needed to schedule Kafka pods to specific nodes. For example because +In order to achieve good performance, it might be necessary to schedule Kafka pods to specific nodes. For example, because they contain special storage hardware or just to run them on dedicated nodes to make sure that Kafka doesn't interfere -with any other applications. To achiave this, it should be possible to configure node selectors for the Strimzi +with any other applications. To achieve this, it should be possible to configure node selectors for the Strimzi deployments. ## Custom Resource Definitions support -Cluster and Topic Controllers should have support for [Custom Resource Definitions (CRDs)](https://kubernetes.io/docs/concepts/api-extension/custom-resources/). +The Cluster and Topic Controllers should have support for [Custom Resource Definitions (CRDs)](https://kubernetes.io/docs/concepts/api-extension/custom-resources/). ## Service broker support -Cluster Controller should be able to work as a [Service Broker](https://www.openservicebrokerapi.org/). +The Cluster Controller should be able to work as a [Service Broker](https://www.openservicebrokerapi.org/). ## Support for automated cluster balancing During the lifecycle of the Kafka cluster it can happen that it becomes unbalanced. Some nodes are hosting very _heavy_ -topics (i.e. busy topics with a lot of traffic) whicle other nodes are idle most of the time hosting less busy topics. -Automated cluster balancer should continuously monitor the cluster state and balance it (re-distribute the topics) when +topics (i.e. busy topics with a lot of traffic) while other nodes are idle most of the time hosting less busy topics. +An automated cluster balancer should continuously monitor the cluster state and balance it (re-distribute the topics) when needed to make sure that the load is optimally distributed across all cluster nodes. ## Kafka updates @@ -62,4 +59,4 @@ Strimzi should make it possible to smoothly handle updates from one Kafka versio ## Integration with other protocols -Allow to access Kafka using different protocols such as HTTP, AMQP or MQTT. \ No newline at end of file +Allow to access Kafka using different protocols such as HTTP, AMQP or MQTT.