HPCC Systems is a platform purpose-built for high-speed data engineering. This article shows how to build HPCC Platform and set up ECL IDE.
Install HPCC Platforms
Download the installation package of HPCC Platforms corresponding to the operating system here. For example, if my system is Ubuntu-22.04 LTS, then download this file:
dpkg -i hpccsystems-platform-community_9.0.8-1jammy_amd64_withsymbols.deb. There may be some errors at this time. Ignore it, please.
sudo apt-get update
sudo apt-get install -f
Verify that the program installed successfully: Run
cd /opt/HPCCSystems. The installation is successful if no errors have occurred and the directory is entered successfully.
Check plugins: Run
Check current service status: Run
sudo /etc/init.d/hpcc-init status
Start HPCC service: Run
sudo /etc/init.d/hpcc-init start
Check the IP address: Run
ip addr show. You can access
IP addr: 8010to login ECL Watch
Install ECL IDE
ECL IDE is an IDE specially designed for ECL programming language, related documents can be found here.
First, download ECL IDE according to the platform, and find ECL IDE and Client Tools here.
When you log in for the first time, the preference interface will pop up, enter the IP obtained in Install HPCC Platforms in the IP address, click Advanced, and leave Attribute Server blank. Other options can be set according to your preferences.
In the ECL Folder of the Compiler tab, add your own code directory.
In the lower right Repository window, in My Files, right click and Insert Folder, then in the newly created folder, right click and Insert File.
In the newly created file, enter:
Click Submit in the upper right corner, and you can see the result in the result tab.
Install the HPCC Systems GNN bundle
Before installing the GNN bundle, you need to install TensorFlow first.
My HPCC Platform is on WSL2-Ubuntu22.04, and my ECL IDE is on Windows, so I need to install GNN Bundle in Windows.
run the below code in cmd:
"C:\Program Files (x86)\HPCCSystems\9.0.8\clienttools\bin\ecl.exe" bundle install https://github.com/hpcc-systems/ML_Core.git
You can test whether the installation is successful by running