diff options
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 19 |
1 files changed, 9 insertions, 10 deletions
@@ -13,13 +13,13 @@ OpenDC is an open-source simulator for datacenters aimed at both research and ed  -Users can construct datacenters (see above) and define experiments to see how these datacenters perform under different workloads and schedulers (see below). +Users can construct datacenters (see above) and define portfolios of scenarios (experiments) to see how these datacenters perform under different workloads and schedulers (see below).  -The simulator is accessible both as a ready-to-use website hosted by Delft University of Technology at [opendc.org](https://opendc.org), and as source code that users can run locally on their own machine. +The simulator is accessible both as a ready-to-use website hosted by us at [opendc.org](https://opendc.org), and as source code that users can run locally on their own machine, through Docker. -OpenDC is a project by the [@Large Research Group](http://atlarge-research.com). +OpenDC is a project by the [@Large Research Group](https://atlarge-research.com). ## Architecture @@ -29,13 +29,13 @@ OpenDC consists of four components: a Kotlin [simulator](/simulator), a MongoDB <img src="misc/artwork/opendc-component-diagram.png" alt="OpenDC Component Diagram"> </p> -On the frontend, users can construct a topology by specifying a datacenter's rooms, racks and machines, and create experiments to see how a workload trace runs on that topology. The frontend communicates with the web server over SocketIO, through a custom REST request/response layer. For example, the frontend might make a `GET` request to `/api/v1/users/{userId}`, but this request is completed via SocketIO, not plain HTTP requests. Note that the API itself can also be accessed by HTTP. +On the frontend, users can construct a topology by specifying a datacenter's rooms, racks and machines, and create scenarios to see how a workload trace runs on that topology. The frontend communicates with the web server over SocketIO, through a custom REST request/response layer. For example, the frontend might make a `GET` request to `/api/v1/users/{userId}`, but this request is completed via SocketIO, not plain HTTP requests. Note that the API itself can also be accessed by HTTP. The (Swagger/OpenAPI compliant) API spec specifies what requests the frontend can make to the web server. To view this specification, go to the [Swagger Editor](https://editor.swagger.io/) and paste in our [opendc-api-spec.yml](opendc-api-spec.yml). -The web server receives API requests and processes them in the database. When the frontend requests to run a new experiment, the web server adds it to the `experiments` table in the database and sets is `state` as `QUEUED`. +The web server receives API requests and processes them in the database. When the frontend requests to run a new scenario, the web server adds it to the `scenarios` collection in the database and sets its `state` as `QUEUED`. -The simulator monitors the database for `QUEUED` experiments, and simulates them as they are submitted. It writes the resulting `machine_states` to the database, which the frontend can then again retrieve via the web server. +The simulator monitors the database for `QUEUED` scenarios, and simulates them as they are submitted. It writes the results to parquet files, which in turn are analysed and aggregated by a Spark pipeline. This pipeline then write the aggregated summary to the database, which the frontend can then again retrieve via the web server. ## Setup @@ -69,7 +69,7 @@ git clone https://github.com/atlarge-research/opendc.git cd opendc/ ``` -In the directory you just entered, you need to set up a set of environment variables. To do this, create a file called `.env` in the `opendc` folder. In this file, replace `your-google-oauth-client-id` with your `client_id` from the OAuth client ID you created and put the path of this `opendc` folder in the place of `/your/path/to/opendc`. For a standard setup, you can leave the other settings as-is. +In the directory you just entered, you need to set up a set of environment variables. To do this, create a file called `.env` in the `opendc` folder. In this file, replace `your-google-oauth-client-id` with your `client_id` from the OAuth client ID you created and put the path of this `opendc` folder where this placeholder is currently listed: `/your/path/to/opendc`. For a standard setup, you can leave the other settings as-is. ```.env MONGO_INITDB_ROOT_USERNAME=root @@ -84,8 +84,7 @@ OPENDC_ROOT_DIR=/your/path/to/opendc OPENDC_SERVER_BASE_URL=http://localhost:8081 ``` -Afterwards, you should also create a `traces/` directory in which you place the VM and workflow traces you want to -experiment with. +We provide a list of default traces for you to experiment with. If you want to add others, place them in the `traces` directory and add entries to the database (see also [the database folder](database/mongo-init-opendc-db.sh)) If you plan to publicly deploy, please also tweak the other settings. In that case, also check the `docker-compose.yml` for further instructions. @@ -99,4 +98,4 @@ docker-compose build docker-compose up ``` -Wait a few seconds and open `http://localhost:8081` in your browser to use OpenDC. +Wait a few seconds and open `http://localhost:8081` in your browser to use OpenDC. We recommend Google Chrome for the best development experience. |