Florida-based Stein Mart, a large department store chain, wanted faster data reporting and new analysis capabilities. It switched from an IBM-based business intelligence system to a solution based on Microsoft SQL Server 2008 built on HP technology platform. As a result, Stein Mart runs weekly sales reports in 3 hours instead of 14 and has reduced its monthly technology costs by U.S.$50,000. The company also has new tools for business analysis and forecasting. Business Needs
Founded in Jacksonville, Florida, in 1902, Stein Mart is a department store chain with 268 stores in the United States. The stores specialize in brand-name clothing, accessories, shoes, and home decor at prices up to 60 percent lower than competitors.
||Querying data is much faster than before with the new solution based on SQL Server 2008 and SQL Server Fast Track Data Warehouse. Now, our executives can access the latest business data on time.
Data Systems Manager
As a player in the competitive retail industry, Stein Mart relies heavily on up-to-date data analysis and reporting. For example, company executives meet regularly to discuss the latest sales and inventory reports. “They compare planned versus actual sales performance and look at what’s working or not working in terms of in-store promotions,” says Ilan Wajsman, Data Systems Manager, Stein Mart. “It’s critical that they receive this data promptly for each planning session.”
However, the company’s data warehouse system, which was based on IBM DB2 and AS/400 data servers, had performance issues that often led to data reporting delays. “There were database column and row limitations on the AS/400, so we could only keep a certain amount of data,” says Wajsman. “Also, because of performance issues, we couldn’t bring our transaction log data into the data warehouse quickly enough.” As a result, weekly data processing took at least 14 hours. Wajsman says, “On a regular basis, the data wasn’t ready for planning sessions.”
Stein Mart also needed to improve its business analysis capabilities. “Our business revolves around merchandise reports that compare sales data, but we really wanted to become more analytical and be able to do data mining, sales forecasting, and analysis,” says Wajsman.
Cost was another major issue. The company was spending thousands of dollars a month in hardware, licensing, and maintenance costs to support its AS/400 data warehouse. Solution
In August 2009, Stein Mart began the process of evaluating business intelligence offerings. After looking at technologies from Oracle, Netezza, and Microsoft, the company decided on a new Microsoft solution based on Microsoft SQL Server 2008 Enterprise database software, SQL Server 2008 Reporting Services, and SQL Server Fast Track Data Warehouse. The solution also includes Excel Services in Microsoft Office SharePoint Server 2007.
SQL Server Fast Track Data Warehouse is a set of reference architectures for data warehousing that uses pretested HP technology and helps increase SQL Server 2008 scalability to up to 48 terabytes. The solution uses balanced configurations that can provide up to 200 MB per second per processing unit core.
Stein Mart worked with Microsoft Gold Certified Partner Pragmatic Works to migrate the AS/400 environment to SQL Server 2008 and to convert data reports from the old system to the new system. Pragmatic Works deployed the solution on an HP DL785 G6 server and an HP StorageWorks MSA2000fc G2 storage area network.
Now, sales data is loaded sequentially into partitions in SQL Server 2008 using parallel loads. The solution also uses data compression to shrink the company’s multi-terabyte data warehouse to less than 2 terabytes, which helps increase data query performance.
Using the new solution, Stein Mart produces sales reports much faster than before and expects to improve its market analysis and business forecasting. The company will also reduce its system maintenance costs by $50,000 every month.
Speeds Reporting Processes
Stein Mart can now load data sequentially, which means data is loaded faster, and larger data amounts can be queried. As a result, weekly sales reports are processed in less than 3 hours, compared to 14 hours with the previous system. “Querying data is much faster than before with the new solution based on SQL Server 2008 and SQL Server Fast Track Data Warehouse,” says Wajsman. “Now, our executives can access the latest business data on time.”
Improves Analysis and Forecasting
With its new solution, Stein Mart now has the ability to pull transactional data and analyze its business from end to end. Using SQL Server 2008 Reporting Services, for instance, the company can perform in-depth business analysis of its transactional database. “With SQL Server 2008 and SQL Server Fast Track Data Warehouse, we can look at our entire transaction history and do market analysis and business forecasting, which we could not do previously,” says Wajsman. “Now, for example, we will be able to study the correlation between promotions and sales.”
These capabilities make it possible for Stein Mart to use data to demonstrate business success. “This technology gives us the ability to be more fact-based as a company,” says Wajsman. “We can make business decisions based on demonstrated sales history.”
Saves $50,000 in Monthly Costs
By implementing the Microsoft solution, Stein Mart was able to consolidate its AS/400 data warehouse into one commodity-based hardware server, which contributes to a monthly savings of $50,000 in hardware leasing and maintenance costs. “We will be able to eliminate 95 percent of the monthly costs related to the previous system,” says Wajsman.
Also, Stein Mart IT employees will no longer need to spend their time managing the AS/400 servers. “Maintaining that system required multiple employees with diverse skill sets. We need fewer people to operate the new system,” says Wajsman. “The time savings helps us focus on business and customer challenges, rather than technical ones.”This case study is for informational purposes only. MICROSOFT MAKES NO WARRANTIES, EXPRESS OR IMPLIED, IN THIS SUMMARY.