MatchUp Hub:Establishing Benchmarks

From Melissa Data Wiki
Revision as of 23:36, 24 September 2018 by Admin (talk | contribs) (Created page with "← MatchUp Hub ==Migrating from Proof of Concept to Production Deployment== The purpose of the establishing a benchmark – before proceeding to live prod...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

← MatchUp Hub

Migrating from Proof of Concept to Production Deployment

The purpose of the establishing a benchmark – before proceeding to live production data and custom matchcodes is to establish that the said requirements will be possible in your environmment.

MCO Benchmark ChartStep1.png

For this reason, we have provided a 1 million record benchmarking file to establish that your environment will most likely not cause slow performance when moving forward. In each benchmarking case.

The benchmark should be established running the sample data local with respect to the local installation, use the recommended default matchcode, and should not use available advanced options.

Get Benchmarking Files

For MatchUp Object, we have provided sample benchmarking scripts. For ETL solutions, benchmarks should use the following configured settings, which simulate the most basic usage of the underlying object. And also represents the first step in a Proof of Concept to Production migration….

MCO Benchmark ChartStep2.png


A simple file read, MatchUp keybuilding, deduping and output result stream will be established as the lone Data Flow operation…

MCO Benchmark DataFlow.png

All distributions of MatchUp provide the benchmark matchcode…

MCO Benchmark Matchcode.png

Basic input Field Mapping should be checked…

MCO Benchmark FieldMapping.png

Force generation of important output properties compiled during a MatchUp process, although we’ll evaluate only the Result Code property to get Total record and Duplicate counts…

MCO Benchmark Options.png

Although configuring Pass-Through options for all source data fields should have a negligible effect on our benchmarking, it should be noted that in actual production, Passing a large amount of source data and or Advanced Survivorship can slow down a process considerably….

MCO Benchmark PassThrough.png

We’ll create two output streams that will allow us to tally the Total and Duplicate counts….

MCO Benchmark Output.png

After the benchmark program has been configured and successfully run, you are ready to make small incremental steps…

MCO Benchmark ChartStep3.png

If returned benchmark results do not closely resemble the expected benchmarks provided by Melissa Data, please fill and return the Benchmark form and return to Melissa Data. Completeness and additional comments provided can help eliminate current or potential performance issues when the process is scaled up to resemble a production process.