Of course you can't -- what a stupid question. However, asking yourself how you could save money by using systems that are optimized to run a specific type of business workload is a very smart and often overlooked question. Transactional applications put considerably different demands on a computing infrastructure than do analytics workloads, for example. Therefore, how you structure data and allocate processors, memory, and storage has a great impact on overall performance and costs.
When it comes to IT environments, we all have generally trusted the old adage: “It is not broken, don’t fix it.” Recently, two opposing forces have caused us to broaden the definition of broken to include inefficient. Those forces are: 1) the growing need to improve business results by analyzing more information faster than ever before; and 2) the perpetual need to reduce costs. The only way to satisfy both these demands is to improve the cost efficiency of existing systems, and shift the resources saved to the new projects needed to drive growth.
The use of general purpose systems to run all types of application workloads can lead to inefficiencies in a number of ways. You could use the systems as is and fail to deliver optimal performance. As a result, more system resources (servers, storage, software, space, administration, power, and cooling) are required to achieve the level of service required of the business. Or, general purpose systems can be integrated and tuned for optimal performance -- which costs time and expense of experts on your staff or the professional services you hire. Alternatively, you can use systems already optimized by experts for specific workloads that can save you from wasting valuable system resources and free your staff to do new work that adds business value.
For example, businesses using DB2 to manage SAP application data are taking advantage of a decade of optimizations built in by IBM and SAP engineers to save 30 percent or more on costs, while also seeing performance improvements typically in the range of 20-40 percent. Businesses that are working to extract new value from the growing volume of data from smart meters and sensors are using Informix TimeSeries data management to save more than 50 percent on storage space, and cut data load and analysis times from several hours to minutes, vs. conventional relational database systems. Others are using Netezza analytic data warehouse appliances to deploy solutions that unlock new insights in days and without expert tuning. One Netezza client noted the ability to deploy more projects in three months than they had been able to in three years with a general purpose database system that required a great deal of expert tuning and management for deep analytic workloads.
These are obviously just a few quick examples from the part of the IBM portfolio that keeps me busy every day. But I believe they demonstrate my general point. A lot of time and money can be saved by paying attention to the efficiency of your systems -- and by taking advantage of systems that have already been optimized by experts for the specific workload you are running.
— Bernie Spang, Director, Strategy and Marketing, IBM Database Software and Systems