I am interested in KNIME Server, but there are a few topics which are puzzling me. You can see these questions below:
In every server editions (small, medium and large) there is an expression which is 5 users and 8 cores. Can you explain that expression? What does it mean?
What is pushdown? Can KNIME Server and KNIME Analytics Platform do pushdown?
Thanks in advance,
- These terms are explained at the bottom of this page: https://www.knime.com/knime-software/knime-server-pricing
User : a user is an individual person with a unique ID and may only be an employee, contractor, advisor, or agent of the customer who is authorized by the customer to use KNIME Software. Users have access to the full functionality of KNIME Software and are active for as long as they own items on KNIME Server such as workflows, data, folders, or scheduled jobs, etc. The license defines the permitted number of users.
Core: a core refers to a physical CPU core in the case of physical hardware and a “virtual CPU” core in the case of infrastructure as a service. In cases where the infrastructure as a service defines a relationship between physical CPUs and virtual CPUs, the physical CPU core count shall prevail. The number of licensed cores refers to the maximum number of CPU cores that can be allocated to running the software. The license defines how many CPU cores can be used by KNIME Software.
- KNIME supports pushdown execution. The Database Manipulation nodes push(back) SQL. For example for joining you should use Database Joiner node which will not read your tables into KNIME. With our KNIME Big Data nodes you can also push down the execution to the cluster. https://www.knime.com/knime-big-data-extensions
Thank you @cpadilla for your answers. There is one more question to be clearified for me;
How can i metadata and lineage analysis on KNIME Server by using metadata repository? For instance, how many workflow using x column of y table? or Is a table, generated by workflow A, using workflow B as a source?
Thanks in advance,
With KNIME 4.2 you can extract all workflow metadata into JSON or XML so it can be used for reporting. This will give you nodes being used, tables, and values. You can then use these to create a report or build statistics. https://www.knime.com/metadata-mapping?utm_source=pr&utm_medium=wire&utm_campaign=release20
Here is an example on KNIME Hub: https://hub.knime.com/knime/spaces/Examples/latest/09_Enterprise/03_Metadata_Mapping/Generate_Workflow_Metadata_Overview_Report