Eric M Russell (1/12/2015)

---

Using hash or checksum techniques to detect modifications to bulk numbers of records is inefficient (it has to read and hash all columns) and not reliable (data...

TheSQLGuru (1/11/2015)

---

Most of my healthcare work has been directly for services providers or insurers, although I have worked directly with Centricity and a few others, especially on data movement stuff....

Jeff Moden (1/11/2015)

---

Ah... understood on the OR.

I'm thinking that a UNIQUE NIX on the ID and Hash (and that combination should be unique because the ID is unique) might be...

Jeff Moden (1/11/2015)

---

bantrim (1/11/2015)

---

Jeff Moden (1/11/2015)

---

It may be that having the clustered index on the hash would work faster on the lookups, especially if it were a unique index that...

Okay, if anyone reading this is interested, the results are in.

The benchmarked table held just over 53 million records, and contained 37 columns, totaling around 10 GB total space. ...

Jeff Moden (1/11/2015)

---

It may be that having the clustered index on the hash would work faster on the lookups, especially if it were a unique index that was guaranteed to...

TheSQLGuru (1/11/2015)

---

1) you REALLY need to read Thomas Kejser's blog post on hashing in SQL Server: http://kejser.org/exploring-hash-functions-in-sql-server/. I believe he has others.

2) I have used what you are doing...

TheSQLGuru (1/11/2015)

---

1) you REALLY need to read Thomas Kejser's blog post on hashing in SQL Server: http://kejser.org/exploring-hash-functions-in-sql-server/. I believe he has others.

2) I have used what you are doing...

Jeff Moden (1/11/2015)

---

It may be that having the clustered index on the hash would work faster on the lookups, especially if it were a unique index that was guaranteed to...

Jeff,

Thank you, both for your input, as well as your complements. To answer your questions:

1.) The total size of all tables comes out to around 900 GB over...

I'll second the PSEXEC suggestion. I...

MMartin1 (12/31/2014)

---

One of the tweaks I do is alter the max rows per batch and or commit size when transferring millions or rows with SSIS's, dataflow's ole db destination. This...

MMartin1 (12/31/2014)

---

Do you have access to SSIS? You could query the data with the use of a data source component, of the dataflow task. It is much faster than a...

Jeff Moden (12/28/2014)

---

I do understand about not being able to see anything of what you need. The full table pulls using BCP probably are the best way to go...

Jeff Moden (12/28/2014)

---

bantrim (12/28/2014)

---

Jeff,

Thank you for your reply. I haven't tested the queryout option against the incremental query, but I guess I had assumed that I would run into...