Difference between revisions of "Sahara/WhyNotHeat"
Line 3: | Line 3: | ||
Generally using Heat underneath for infrastructure provisioning looks reasonable. In a tactic perspective there are few factors making Heat usage underneath Savanna problematic: | Generally using Heat underneath for infrastructure provisioning looks reasonable. In a tactic perspective there are few factors making Heat usage underneath Savanna problematic: | ||
* Heat stability for Grizzly release. Savanna currently maintains Grizzly+ compatibility. | * Heat stability for Grizzly release. Savanna currently maintains Grizzly+ compatibility. | ||
+ | Heat Solution: stability of main functionality should be a lesser issue after Havanna release | ||
* Installation of large Hadoop clusters (100+ nodes). Will be addressed by proposed architecture changes. | * Installation of large Hadoop clusters (100+ nodes). Will be addressed by proposed architecture changes. | ||
Heat Solution: the following feature: https://blueprints.launchpad.net/heat/+spec/multiple-engines will be implemented in early Icehouse. That will solve problem of bringing up big clusters. | Heat Solution: the following feature: https://blueprints.launchpad.net/heat/+spec/multiple-engines will be implemented in early Icehouse. That will solve problem of bringing up big clusters. | ||
* Anti-affinity support for HDFS redundancy in cloud environment | * Anti-affinity support for HDFS redundancy in cloud environment | ||
− | Heat Solution: after instance node groups (https://blueprints.launchpad.net/nova/+spec/instance-group-api-extension) are implemented in Icehouse, we | + | Heat Solution: after instance node groups (https://blueprints.launchpad.net/nova/+spec/instance-group-api-extension) are implemented in Icehouse, we can add their support into Heat |
* Circular dependencies - we should generate ‘/etc/hosts’ for all instances in provisioned cluster. We can’t use cloud init for this directly. There are a couple possible solutions using Heat, but none of them looks like a straightforward solution. | * Circular dependencies - we should generate ‘/etc/hosts’ for all instances in provisioned cluster. We can’t use cloud init for this directly. There are a couple possible solutions using Heat, but none of them looks like a straightforward solution. | ||
Heat Solution: run a combination of os-collect-config and os-apply-config to generate /etc/hosts | Heat Solution: run a combination of os-collect-config and os-apply-config to generate /etc/hosts |
Revision as of 15:44, 24 September 2013
1. Why doesn’t Savanna use Heat to provision VMs?
Generally using Heat underneath for infrastructure provisioning looks reasonable. In a tactic perspective there are few factors making Heat usage underneath Savanna problematic:
- Heat stability for Grizzly release. Savanna currently maintains Grizzly+ compatibility.
Heat Solution: stability of main functionality should be a lesser issue after Havanna release
- Installation of large Hadoop clusters (100+ nodes). Will be addressed by proposed architecture changes.
Heat Solution: the following feature: https://blueprints.launchpad.net/heat/+spec/multiple-engines will be implemented in early Icehouse. That will solve problem of bringing up big clusters.
- Anti-affinity support for HDFS redundancy in cloud environment
Heat Solution: after instance node groups (https://blueprints.launchpad.net/nova/+spec/instance-group-api-extension) are implemented in Icehouse, we can add their support into Heat
- Circular dependencies - we should generate ‘/etc/hosts’ for all instances in provisioned cluster. We can’t use cloud init for this directly. There are a couple possible solutions using Heat, but none of them looks like a straightforward solution.
Heat Solution: run a combination of os-collect-config and os-apply-config to generate /etc/hosts
- Level of complexity. We try to keep things as simple as possible. Adding extra layer will increase overall complexity of the solution. In addition both Savanna and Heat under active development changing lots of internals and even APIs and will require extra effort to coordinate.
Once Heat supports all the features we need, it will make sense to use Heat to provision VMs for Savanna. Here is what we’ll do: - Create a wiki page with text from this email - Create a list of requirements for Heat
Once Heat fulfills all these requirements we will be able and should use Heat for VM provisioning.
2. Why we need Savanna? Can’t we use Heat to do what Savanna does?
- Savanna provides bunch of Hadoop-specific features. It’ll be hard to provide them as Heat plugin
- Savanna provides Hadoop-specific APIs and functionality. Heat use cases are mostly around provisioning/deployment.
- Savanna provides integration with various Hadoop distributions through pluggable mechanism
Now, more details on each item. Hadoop specific features:
- Tight Swift integration. Hadoop can read and write from/to Swift object storage. Savanna provides required configs for the Hadoop cluster.
- Usage of anti-affinity to preserve data-redundancy of HDFS nodes
Hadoop-specific APIs and functionality:
- Hadoop cluster scaling
- Elastic Data Processing: https://wiki.openstack.org/wiki/Savanna/EDP
Integration with Hadoop distributions through pluggable mechanism:
Usually Hadoop cluster deployment is a multi-step operation. First step is to install management console (for instance Apache Ambari). Second step is to communicate with management console through REST API to provision Hadoop on the cluster. Savanna wraps all this operations under well-defined API.
3. Why can’t Savanna be used as a plugin for Heat?
It should be and it will be someday.