Showing posts with label boss. Show all posts
Showing posts with label boss. Show all posts

Friday, March 30, 2012

Question About Scalability of Reporting Services

We use SQL Server 2000 Reporting Services to represent reports for our company.

Our boss complained to me about the poor performance of Reporting Services when he viewed reports. He suspected that RS Server is a single-thread server and can not be scaled up by adding more CPUs or memories to increase the reports process speed.

Is my boss correct?

And how can we increase the performance of Reporting Servces?

Traditional scale up can resolve the issue?

Appreciate your precious reply.

Your boss is not correct :)

We will scale up with more CPUs. Most of Reporting Services is managed code, so we are subject to whatever memory constraints are in place by the CLR. So on 32-bit machines we will use up to 2GB of memory, and we will use more on 64-bit platforms.

You can check the execution log to see which reports in particular are taking a long time to execute and what phases of those executions are long running:

http://msdn2.microsoft.com/en-us/library/ms159110.aspx

See Tudor's blog entry on performance here as well:

http://blogs.msdn.com/tudortr/archive/2004/06/28/167969.aspx

It has good tips for optimizing a poorly performing report.

|||

Excellent!

Thanks alot!

Friday, March 23, 2012

Question about normalization

I have a question about normalization
basically I have an excel file with the following columns that my boss wants to store in a small database.


Brand Name, Retail Store, Location Info

Each Retail store may have one or more address. Each Retail store handles one or more brands Some brands may NOT be available in some Locations

So basically I created the following entities.

    Brand (BrandID, Brand) Company (CompanyID, Name, Type (retail, manufacturer, etc)) Cmp_Location (LocationID, Address info..., CompanyID) Location_Brand (LocationID, BrandID)

What do you guys think?

You've laid out a decent start.

I wonder if, however, it might be advantageous to Consider that a Brand may have multiple Products. Perhaps a BrandProduct table will prove useful. (And then of course, there may eventually be a need for a BrandProductsDetail table.)

The Retail Stores carry Products, and they may not carry ALL Products for a Brand.

As a minor point, I would name the Locations table something like: CompanyLocations -it will sort following the Company table in the event your project continues to grow and develop a need for more tables.

(And with large retail operations, Brand is just a sub-component of a Supplier. One Supplier may control multiple Brands.)

If you are using SQL 2005. refer to Books Online about the use of schemas. You could have a schema for Suppliers, one for Retailers, etc. That would make it both easier to use Table names that are meaningful without being complex AND keep them located together. Consider

Suppliers.Companys Suppliers.Products Suppliers.ProductDetails Suppliers.Locations Retailers.Companys Retailers.Locations Retailers.Products