M query - Cell values

Streaming datasets

PowerQuery Diagnostics

Config Premium Capacity

Managing Premium Capacity

Optimizing capacity

PBI REST API connection

XMLA Basics

XMLA - MS Doc

Partition Refresh using XMLA

Desktop Performance Analyzer

Query Diagnostics

Monitoring Query

Optimizing Power Query

Aggregations and debug

Large Tables

RLS

OLS

Automatic page refresh Custom Visual Governance

Points:

1. Maximum concurrent refreshes - 1.5 times the v-core capacity rounded up to nearest integer
2. Automatic Page refresh - Setting present in admin side overrides the refresh setting per page and query running time is the actual page refresh frequency
3. For a full refresh, twice the amount of memory is required
4. While doing OLS, take relationships into consideration. Hiding a single table with relation to other tables results in error
5. Calculated columns in DAX is a single thread operation and is also non-parallel process. More number of calculated columns is going to make the refresh longer.
6. Vertipaq stores data in columnar format and Storage Engine (SE) processes them for the datacache. SE scans only the columns necessary for the DAX code to work.
7. SE can use parallel operations to form datacache, but FE only uses single thread for its operations.

DAX:

1. SUMX(KEEPFILTERS(),MEASURE) -> Forces to use the filter in KEEPFILTER while evaluating the MEASURE, and the Implicit CALCULATE won't overwrite it during context transition.
2. Nested iterators - Cardinality of the the outer iteration defines the peformance.
3. USERELATIONSHIP cannot be used when row level security is defined for the table in which the measure is included.