Today's leading databases cannot handle many common queries and should only be used as a last resort, claims a Bloor Research report released this week.
The analyst company identified a variety of increasingly common database queries that were particularly poorly handled, and recommended that users turn to specialist third-party products instead.
Author of Innovations in Data Warehousing - Answering complex, real-time and dynamic queries. An Evaluation and Comparison, Philip Howard, argues that a conventional access approach as implemented in Oracle, IBM DB2 and Microsoft SQL Server should only be used as a last resort.
"Most [enterprises] are going to have a relational data warehouse that is augmented with online analytical processing (Olap) for slice and dice. But certain types of queries don't fit in, such as relatively complex queries working on transactional data," he said.
Among those queries identified were: unpredictable and dynamically changing queries for which the database could not be optimised, those involving large table scans, real-time queries (most are near real-time at best), and time-lapse queries such as 'who bought a barbecue within seven days of buying patio furniture?'.
In addition, queries involving qualitative alongside quantitative information needed to include textual access methods.
Six specialist products were evaluated as part of the report - Aleri, Alterian, Apama, Aruna Companion, Sand Nucleus and Sybase IQM - and results were compared with those for a conventional relational database management system (RDBMS).
The Bloor study reported that Alterian and Aruna managed eight out of 10 unpredictable queries, Aruna nine for queries involving multiple and complex table joins, and Sand nine for large table scans. The RDBMS scored only three in each case.
"In the last decade Olap grew and is now ubiquitous. In this decade I see significant growth for these types of product," said Howard.
But he did not envisage much impact from new 'post-relational' databases that attempt to address relational weaknesses. "The marketing clout [of the leading relational players] is too strong," he said.
Different techniques are used to achieve fast throughput, such as storing and retrieving data by columns to avoid the need to read rows. By contrast, business intelligence tools are constrained by the speed of the relational database.
A summary of the report is available here.
Use the same password for every website? It might be time to change them all
Applicants for parking bay suspensions put at risk of credit card fraud by Islington Council
Robert Swan appointed interim CEO after Brian Krzanich's departure
Should you link your data sets to add value, or leave them separate to reduce risk?