ltu, you wrote
|
2 billion distinct values for a row.
|
Small mistake. This should be (from section A.1 "QlikView Manual Reference"):
|
However, there is one inherent limitation in QlikView that you must be aware of when building very large documents: A QlikView document cannot have more than 2,147,483,648 distinct values in one field.
|
Secondly, you wrote:
|
The functional answer is that you're far more likely to experience hardware limitations before you experience degraded performance unless you have/are planning a server on the order of a TB of RAM and quite a few cores and even then it's more apt to be due to data model and calculations rather than a pure data handling limitation.
|
My Customer has physical server with: 1TB of RAM, 4x10 Cores of Intel Xeon E7 4870 2.4GHz. And he observed QlikView Engine performance degradation in application with data model which takes approximately 140GB of memory (more than 2 billion records, pure data model with data, without cache). Performance degradation is so large that expand list from multibox takes about two minutes, which normally takes less than a second. CPU has a consumption of only a single core during expanding list from multibox. I have an open case with the QT Support, the case number: 502034.
In this thread I'm looking for Partners or Customers with a HUGE app (more than 140GB in RAM) I am looking for Qlik Partners or Qlik Customers who have experience dealing with HUGE applications, which takes more than 140GB in RAM. Does anybody have one?
BR
TH
