4ebff7618e
The patch improves the benchmark config format in that it moves some technical
parameters for each scenario run (like "times", "tenants" etc.) in a special
"config" section and also renames the "concurrent" parameter to "active_users",
which is way more clear for the end-user.
A typical benchmark config now would look like this:
"benchmark": {
"NovaServers.boot_and_delete_server": [
{"args": {"flavor_id": 1,
"image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
"execution": "continuous",
"config": {"times": 10, "active_users": 2,
"tenants": 3, "users_per_tenant": 2}}
]
}
The new mandatory "execution" parameter indicates the means of scenario
execution and can now be set to only one value, namely "continuous". Another
execution type, "periodic", will be implemented in another patch.
The patch also improves unittest coverage for possible benchmark config issues
and replaces some nested mock calls with a better wrapper syntax.
Blueprint flexible-benchmark-config
Change-Id: Iabc90838aa8ad9cb44d73892852ed21c652ad274
Rally
Introduction
Rally is a Benchmark-as-a-Service project for OpenStack.
Rally is intended to provide the community with a benchmarking tool that is capable of performing specific, complicated and reproducible test cases on real deployment scenarios.
In the OpenStack ecosystem there are currently several tools that are helpful in carrying out the benchmarking process for an OpenStack deployment. To name a few, there are DevStack and FUEL, which are intended for deploying and managing OpenStack clouds, the Tempest testing framework which validates OpenStack APIs, some tracing facilities like Tomograph with Zipkin. The challenge, however, is to compile all these tools together on a reproducible basis. That can be a rather difficult task since the number of compute nodes in a practical deployment can easily be large and also because one may be willing to use many different deployment strategies that pursue different goals (e.g., while benchmarking the Nova Scheduler, one usually does not care of virtualization details, but is more concerned with the infrastructure topologies; while in other specific cases it may be the virtualization technology that matters). What Rally aims to do is Compile many already existing benchmarking facilities into one project, making it flexible to user requirements and ensuring the reproducibility of test results.
Architecture
Rally is split into 4 main components:
Deployment Engine, which is responsible for processing and deploying VM images (using DevStack or FUEL according to user’s preferences). The engine can do one of the following:
- deploy an Operating System (OS) on already existing VMs;
- starting VMs from a VM image with pre-installed OS and OpenStack;
- delpoying multiple VMs inside each OpenStack compute node based on a VM image.
VM Provider, which interacts with cloud provider-specific interfaces to load and destroy VM images;
Benchmarking Tool, which carries out the benchmarking process in several stages:
- runs Tempest tests, reduced to 5-minute length (to save the usually expensive computing time);
- runs the user-defined test scenarios (using the Rally testing framework);
- collects all the test results and processes the by Zipkin tracer;
- puts together a benchmarking report and stores it on the machine Rally was lauched on.
Orchestrator, which is the central component of the system. It uses the Deployment Engine, to run control and compute nodes, in addition to launching an OpenStack distribution. After that, it calls the Benchmarking Tool to start the benchmarking process.
Links
Wiki page:
Launchpad page:
Code is hosted on github: