Abstract One of the greatest needs in offshore geotechnical engineering is for large scale test measurements on which to calibrate design procedures. The geotechnical centrifuge offers at least a partial remedy. Because it allows one to properly simulate stresses, it is a legitimate, relatively inexpensive option to full scale field testing. As such it is a valuable technique and can be an excellent complement to laboratory tests, l-g model tests and numerical analyses. However, it has not been widely used by industry even though the capability has existed for almost thirty years. This paper argues that this technology should gain acceptance beyond the research community, The paper presents an overview of centrifuge principles, philosophies of use, and limitations of the technique. For illustration, several actual applications of centrifuge testing for complex offshore problems are described. Results are shown to provide important insights into prototype behavior and to agree well with full scale measurements where these are available. Introduction Probably the greatest need in offshore geotechnical engineering is for relevant large scale field measurements. For example, a great deal of effort has been expended in collecting all the axial load tests of piles that might be relevant to offshore applications (Pelletier, et all. This effort has clearly shown that there are relatively few tests that are sufficiently well documented that they can be used in any quantitative analysis and fewer still that are even within an rder of magnitude of the load carrying capacity of modem deep water foundations. This is particularly noteworthy since the design procedures (AP[2) are basically empirical and it is prohibitively expensive to carry out site specific offshore field load tests as is frequently done on land. A few tests, mostly onshore, have been carried out by industry that were especially designed to be applicable to offshore conditions. These few tests play a very important role in calibrating our design procedures but there are clearly major drawbacks in such test programs:the largest piles that have been tested have capacities that are an order of magnitude or more smaller than full scale offshore piles,most tests have beenconducted on pile segments embedded at relevant depths through eased holes,a single program can cost several million dollars to perform,the programs consist of very few tests and hence the test conditions are very limited, andit is virtually impossible to find a site with ideal conditions so that some loss of control is inevitable. At the same time it can be argued that in spite of a fifty year history with offshore piles, reliable observations of their performance are relatively scarce.