## NodeMinRowsProcessed in the Graphs

I am confused about what this means in the graphs.

At one point I was told this is the minimum node that had data and NodeMaxRowsProcessed in the Graphs was the maximum node that had data in the graphs. So if NodeMinRowsProcessed had 2 in it and NodeMaxRowsProcessed had 317 in it for a 400 node cluster nodes 1, and nodes 318 to 400 had no data. That seemed to be the case for the 5.0 graphs.

However, looking at graphing in 6.0 I find that NodeMinRowsProcessed and NodeMaxRowsProcessed seems to mean something different. NodeMinRowsProcessed seems to mean the node where the minimum rows were processed. NodeMasRowsProcessed seems to mean the node where the maximum rows were processed.

I have graphs in 6.0 where the number for NodeMinRowsProcessed = 324 and the NodeMaxRowsProcessed = 106.

So two questions:

1. What does NodeMinRowsProcessed mean in 6.0?

2. Is the best way to determine if the distribution is not working well would be to compare the the number of average rows per node vs the maximum rows processed on a node vs the minimum rows processed? Also to look at the maximum and minimum skew?

I am open to suggestions.

At one point I was told this is the minimum node that had data and NodeMaxRowsProcessed in the Graphs was the maximum node that had data in the graphs. So if NodeMinRowsProcessed had 2 in it and NodeMaxRowsProcessed had 317 in it for a 400 node cluster nodes 1, and nodes 318 to 400 had no data. That seemed to be the case for the 5.0 graphs.

However, looking at graphing in 6.0 I find that NodeMinRowsProcessed and NodeMaxRowsProcessed seems to mean something different. NodeMinRowsProcessed seems to mean the node where the minimum rows were processed. NodeMasRowsProcessed seems to mean the node where the maximum rows were processed.

I have graphs in 6.0 where the number for NodeMinRowsProcessed = 324 and the NodeMaxRowsProcessed = 106.

So two questions:

1. What does NodeMinRowsProcessed mean in 6.0?

2. Is the best way to determine if the distribution is not working well would be to compare the the number of average rows per node vs the maximum rows processed on a node vs the minimum rows processed? Also to look at the maximum and minimum skew?

I am open to suggestions.

- georgeb2d
**Posts:**93**Joined:**Wed Dec 24, 2014 3:36 pm

georgeb2d,

HTH,

Richard

This is similar to the way the skew numbers are evaluated -- the number of records that were processed on the one node that processed the fewest (for NodeMinRowsProcessed) or most (for NodeMaxRowsProcessed) rows of data. They definitely do NOT specify which node that is. I don't have a 5.0 system readily available to verify it, but my assumption is that this has not changed for 6.0 and that is also the way it was in 5.0.1. What does NodeMinRowsProcessed mean in 6.0?

Yes.2. Is the best way to determine if the distribution is not working well would be to compare the the number of average rows per node vs the maximum rows processed on a node vs the minimum rows processed? Also to look at the maximum and minimum skew?

HTH,

Richard

- rtaylor
- Community Advisory Board Member
**Posts:**1601**Joined:**Wed Oct 26, 2011 7:40 pm

2 posts
• Page

**1**of**1**### Who is online

Users browsing this forum: No registered users and 1 guest