Tag Archives: table-valued parameters

Persisting Data: part 9 – A Modest Data Collection Proposal

This begins with a  recap the previous posts in this series. Part 1 is a technology review of some of the more common .NET application layer persistence techniques. One key take-away is that many currently trending object relational mappers (ORMs) … Continue reading

Posted in Data Loading | Tagged , , , | Leave a comment

Persisting Data – part 8: Concurrent Persistence Processing

Contention is the cost of concurrency. SQL Server does a pretty good job of managing contention out of the box. For those times when the requirements get absolutely nutty SQL Server also provides a variety of database server settings and … Continue reading

Posted in Data Loading | Tagged , , , , | Leave a comment

Persisting Data – part 7: Concurrency and Deadlocking

In part 6, the charts examined an information region close to an edge of the useful test window. In those tests, the test results at low row size values were overshadowed by the overhead of SSIS. Not until the row … Continue reading

Posted in Data Loading | Tagged , , | Leave a comment

Persisting Data – part 6: TVPs are for SSISies

By now we may agree that Table-valued Parameters (TVPs) have an undeniable throughput advantage over row-at-a-time approaches for moving data from application buffers to safe storage. We may even be on the same page with regard to the appeal of … Continue reading

Posted in Data Loading | Tagged , , , , | Leave a comment

Persisting Data – part 5: Concurrency and Locking

The table lock is important for balancing the application’s concurrency and throughput requirements during a persistence operation. We can expect the incoming data load to have better throughput characteristics for our process when the table is locked. We can expect … Continue reading

Posted in Data Loading | Tagged , , , , | Leave a comment

Persisting Data – part 4: Upsert and MERGE

In part 3 test results when data is persisted to an un-indexed heap table were compared. The un-indexed heap was ideal for isolating the importĀ  – get data into the database just far enough so it is safe – operation … Continue reading

Posted in Data Loading | Tagged , , , , | Leave a comment

Persisting Data – part 3: Rows, Sets and Files

In part 2 of this series of “Persisting Data” posts, the effects of several database server side configuration options on the throughput capabilities for SQL Server to load rows into an un-indexed heap were examined. Reporting Services line charts depicting … Continue reading

Posted in Data Loading | Tagged , , , , | Leave a comment

Persisting Data – part 1: The “impedance mismatch”

This post is an introduction to a few posts to follow that will describe a testing methodology for designing optimized database synchronization protocols. Synchronization is the operation of persisting data changes at the application layers to the database. The methodology … Continue reading

Posted in Data Loading | Tagged , , | Leave a comment