Skip to content

Latest commit

 

History

History
58 lines (46 loc) · 3.72 KB

README.md

File metadata and controls

58 lines (46 loc) · 3.72 KB

AsterixDB Client

Prerequisites

  • A suitable *nix environment (Linux, OSX)
  • JDK 1.8+
  • Maven 3.1.1 or greater

Steps

  1. Check out the project in a directory via git. Assuming that the path to the directory is $HOME/asterixClient (or any other directory based on your choice), we will refer to this directory as CLIENT_HOME in the rest of this document.
  2. Set CLIENT_HOME as an environment variable on the machine you are running the client from (replace the path with the directory you checked out the project into in the previous step):
> export CLIENT_HOME=$HOME/asterixClient
  1. Go to CLIENT_HOME and build the project's artifacts by executing the following commands:
> cd $CLIENT_HOME
> mvn clean package 

Upon a successful build, a new directory named 'target' will be created under CLIENT_HOME that contains the jar file for the client with its dependencies. 4. The AsterixDB client needs two configuration files under the 'conf' directory to run:

  • conf.json: This is a json file that contains the settings for various parameters that client uses to run. A template for this file should already be available under '$CLIENT_HOME/conf' directory. Here is the list of parameters and their description:

    Parameter Type Description
    cc string Hostname/IP of the machine running the AsterixDB instance's cluster controller
    port integer HTTP api port for queries in the AsterixDB instance
    query.language string The language of the queries in the workload: 'aql' or 'sqlpp'
    iterations integer Number or rounds the full workload needs to be executed by the client
    stats.file string Full path to the file that will contain the test's statistics by the end of the test
    dump.results boolean Whether to dump the results per query after its execution: 'true' or 'false'
    results.file string Full path to the file that will contain the dump of the queries' results

    As an alternative (e.g. when connecting to another system than AsterixDB) it is also possible to replace the single query.language parameter by 2 more generic parameters:

    Parameter Type Description
    path string The path of the URI that will receive the query
    parameter string The name of the HTTP parameter for the query

    So when setting cc to "127.0.0.1", port to "19002", path to "query", and parameter to "statement", the URI for a query "QUERY" will be http://127.0.0.1:19002/query?parameter=QUERY.

  • workload.txt: This file defines a specific read-only workload which is a sequence of queries that will be run, in order, by the client in each iteration of a test. Each line of the file should be a full, absolute path to either a readable file that contains a complete query in 'aql' or 'sqlpp' or a directory that will be recursively visited. (By a complete query, we basically mean it should be exactly the same as the statements that you would type inside the query-box when using the Web-API of AsterixDB).

  1. Once you modified and saved the configuration and workload files (with 'conf.json' and 'workload.txt' under the '$CLIENT_HOME/conf' directory respectively) you can run the AsterixDB client by invoking the 'run.sh' script under the '$CLIENT_HOME/scripts' directory:
> $CLIENT_HOME/scripts/run.sh

As client runs, it shows messages (for tracing its progress) on the screen and once it finishes successfully, the summary report on the test's statistics can be found in the file specified as 'stats.file' in '$CLIENT_HOME/conf.json' (if 'stats.file' is not set by the user, the client writes it into its default location under the '$CLIENT_HOME/output' directory).