Long Hill Consulting, LLC - Information Systems Consulting

Relational Database Management Systems
RDBMS vendors have been building more functionality into their core database engines, with features such as Java stored procedures, object data models, and so on. Along comes the Internet and the need for no-nonsense, fast data access. The main job of a database system is still data storage and retrieval: "Get me data, and get it to me fast. And when you put it away, make sure it doesn't break or get lost.

A key value we provide our clients is to help them determine when, how, and to what extent a database system should be used on a given project.

Long Hill's areas of expertise include:

Spatial databases more...
Data warehouses more...
Databases on the Web
Physical and logical database design
Performance tuning
Oracle, MS SQL Server, full-text search engines, even FileMaker.

Spatial databases
Our experience is specifically with point-spatial databases, which have played a prominent role in geo-enabling the web. Spatial databases manage three types of two-dimensional objects: points, lines, and polygons. Applications like the Internet Yellow Pages, store locators, and some mobile applications deal with locations, which are represented as points in spatial databases. These applications do distance-based proximity searches, for example, to return the ten restaurants closest to your hotel.

By jettisoning the more complicated spatial indexing and geometric algorithms required to effectively manipulate lines and polygons, point-spatial databases can do location-based searching more efficiently than their generalized counterparts. Moreover, point-spatial algorithms can be incorporated into off-the-shelf database and search engine products, thereby geo-enabling existing data and applications that contain location data.

Marty wrote a primer on point-spatial searching, which you can read here. More detail on Long Hill's experience in spatial databases is here.

Data warehouses
A data warehouse is a repository that stores information integrated from one or more sources, and is used for efficient querying and analysis. The data from the multiple sources are standardized to a common data model. Information in the warehouse is stored in a condensed or summarized form to aid efficient storage and retrieval.

The process of extracting, transforming, and loading (ETL) information from numerous legacy systems, each with invariably different data models and incompatible data formats, is the hardest and least glamourous part of building a warehouse.

The Internet Yellow Pages shares at least one trait with a data warehouse. If the data source used to build the online directory is of poor quality - they often are - removing duplicates, cleaning the data, and trying to improve on how businesses are categorized is as messy a job as the equivalent taks in any data warehouse.

Previous: Home | Next: Local search

I've found PivotCMS to be reliable, easy to use, and feature rich. But more significantly, Pivot it is a thoughtful application -- one that's been sensitively designed and extended to meet the needs of both novices and power users. I have complete confidence in Marty's commitment to continuing to provide high-quality, robust product enhancements going forward. The product, and the developer, has earned my trust.
Meg Houston Maker
Director of External Information Services
Dartmouth College Development

PIVOT has proven to be a highly intuitive, flexible, client driven web content management system. It provides the right level of simplified text and graphic management functions while enabling the user's growing knowledge of html to constantly enhance its functionality and power.
Greg S Franklin
President Gregory S. Franklin Associates