It’s no secret that today companies live on their data, and are producing more of it than ever before. And the data is being produced by a myriad of systems, inside and outside of the organization, in a dizzying array of formats.
Data warehousing provides the foundation not only for BI systems, but for many other uses as well. The data warehouse has become the source of consistent, accurate information for the entire enterprise. It is the “go to” source for:
- Data Mining
- Historical Records
- Data Integration
- Consolidation of Multiple Sources
- Additional Sources and External Data
- Web-site Traffic
- Generated Data (e.g. budgets, forecasts, and quotas)
A data warehouse is not simply an accumulation of important data, it has to be managed, cataloged, validated, indexed, and otherwise made suitable to become the rock-solid foundation of corporate information.
The Challenges
The challenges to developing a high performance data warehouse are considerable and range from the very technical (e.g. storage architecture) to the business oriented (e.g. definition of key terms).
Just some of the questions which will arise are:
- Choice of storage platforms. – typical RDBMS or DW specialized system?
- How to manage data independence from originating systems and location?
- Storage of foundation data vs. query data?
- Response times: what is adequate and how to provide this?
- What is the expected mix of workloads between shorter, tactical queries and longer, strategic queries, and how best to manage this?
- Normalized or de-normalized database structures?
- Where do cubes or federated data models play a role?
- What ETL tools and methods should we use?
- How best to manage the flow of data, along with cleansing and validation?
- What summary tables are needed?
- What is the difference between a data warehouse, a datamart, and an operational data store? How do I know which one(s) I need?
- How can we keep the data flowing, while keeping manual intervention and maintenance to a minimum?
- How will we know the data is correct?
The Solution
While the design issues around data warehouses are complex, they don’t need to be overwhelming. Our approach is to begin with the business requirements (current and future), and let business value drive the technical design decisions.
In today’s business environment, few if any companies can consider a three to five year investment in a “big bang” comprehensive data warehouse development cycle. Ironside Group will work with you to develop 30 to 60 day cycles, showing business value at each step of the way.
Furthermore, Ironside Group can deliver a data warehouse that is optimized for BI consumption. This can be the solid foundation on which to build your BI initiative and all other data intensive initiatives for years to come.
To build the foundation on which all further analysis is based, it’s vital the data is complete, validated, and structured for proper access by the BI tools. In addition, the appropriate ETL, data cleansing, and meta data tools and processes need to be identified.
Ironside Group provides you with an end-to-end view of your information flow, from gathering and cleansing, all the way to the boardroom.
Our experienced consultants and data experts can provide you with services such as:
- Data Extraction Expertise
- Technical and Architecture Design
- Database and Data Tool Assessment
- Database Design
- Metadata Services (e.g. management and coordination, documentation, master data management (MDM), etc.)
- Data Cleansing
- “Non-traditional” Data Storage Options (e.g. cube, in-memory database, column oriented database, and other high-performance technologies and techniques)
- Data Storage Suitability Assessment
- Data Readiness Assessment for Business Intelligence
- Data and Process Optimization
- Project Management
