const getInputData = context => {
    return search.load({id: '9256'});
}

This function is still running after three hours. The search has 255,000 results. Does this seem like an expected amount of time?

I guess it depends how much you're doing in map/reduce but any script running for 3 hours sounds excessive

It's not hit the map stage yet

Yeah, no, that's baaad

What does that mean?

the number of search results sounds doable, so im guessing there is too much data in your results to serialize

This is the criteria

You're "returning" the loaded saved search which means NS is running the search and I have no idea how that works, is it running the search and then handling passing each result one at a time or is it doing something janky like get me result 0 then result 1 then result 2 etc, running the search each time. (just as an example, you'd hope not 😄 ). I normally just run the search myself, paged, directly into an array and return that array. I'm speculating but I have definitely worked with that many results with maybe 20 or so columns with no issue. Although I now see your formulas 🤔

@NElliott I tried this

const getInputData = context => {
    const search = search.load({id: '9256'});

    const resultSet = [];
    const pagedData = search.runPaged({ pageSize: 1000 });
    pagedData.pageRanges.forEach(pageRange => {
        const myPage = pagedData.fetch({ index: pageRange.index });
        myPage.data.forEach(result => {
            resultSet.push(result);
        });
    });

    return resultSet;
}

Yeah, sounds like your formulas are hitting it hard 😞

Yeah, I'd guess that's the issue

Disappointing.

I'm not familiar enough with netsuite2.com schema to port this to SQL quickly. This will definitely be delivered late :)

I'd love to help you but unfortunately my SQL-fu is sorely lacking and like you I'm not familiar with the schema.

Just read @battk’s suggestion - that sounds promising.

BTW, it might be more efficient in your map to load the record rather than perform a search for each record individually.

your screenshots suggests you have multiple processors available to you, so you can try making another copy of the script to use the other processors

Even if you use all processors, I'm under the impression that doesn't block other M/Rs from starting

only one processor is used for getInputData, leaving 4 others to do other stuff

It's hit map 😄