Denali Career Highlights
Denali , the Arctic Region Supercomputing Center's CRAY Y-MP, came on-line on Dec. 28, 1992. At the time of its purchase by the University of Alaska Fairbanks as part of a $25 million grant, it was the largest memory supercomputer in the world. Denali is an eight-processor CRAY Y-MP M98/81024 large memory vector parallel supercomputer with eight gigabytes (GB) of main memory and 89GB of disk storage. The Y-MP provides peak potential parallel performance of 2.6 billion floating-point operations per second. During its five-plus years of service at ARSC it has been available to users more than 97 percent of the time (most of the down time was due to regularly scheduled maintenance); utilization of the system has averaged better than 91 percent. One-thousand-ninety individuals have performed computations on the Y-MP, from 73 different universities, businesses, federal and state agencies; and representing three countries. Although Denali was a state-of-the-art supercomputer when it was new, after six years of major advances in the fast-moving high performance computing arena, it has been surpassed by new technology. It will take two supercomputers to replace it: the CRAY T3E named Yukon , which has been at ARSC since last March; and a CRAY J932 named Chilkoot , which will be available to users in late April. During its years of service at ARSC, researchers used the extremely large memory vector processing capability of Denali to solve all kinds of tough problems. A few highlights include:
Climate Studies Better Understanding of the Effects of Climate and Weather in Alaska - Arctic temperature trends from 1961-1990 show that over the past 30 years Alaska is one of the hot spots in global warming. Researchers developed models that take into consideration the advent of global warming, and scientists are able to predict what the climate in Alaska might be like many years in the future. These predictions are based on the results of numerical models which were run on Denali . Arctic Biology Advancements in Understanding Fish Populations - Computer models of the physical factors that influence fish survival in the Gulf of Alaska are complex. This region includes unique physical factors posing special challenges to numerical modeling. Oceanographers at the NOAA Pacific Marine Environmental Laboratory used Denali to simulate how ocean currents affect the survival of pollock eggs and larvae. Engineering Studies of Highway Properties Over Permafrost - Springtime load restrictions on Alaska roads increase the cost of goods transported via truck by up to hundreds of thousands of dollars every day the restrictions apply. Assessment of road conditions using non-destructive test data analyzed on Denali allowed better understanding of the "weak" pavement sections, and added to engineers' understanding of the behavior of road beds during periods of thaw. This led to reducing or eliminating, in many cases, the need for load restrictions, resulting in substantial savings for the trucking industry and consumers. Earth Sciences Simulations of Underground Hydrology in Permafrost Conditions - For decades, underground hydrology has been studied in more temperate regions but only recently have researchers looked at how permafrost affects ground water flows. Discontinuous permafrost directs subsurface water flows in complicated but predictable patterns. Researchers at the UAF Water Research Center examined subsurface flows and ground water contamination using Denali . This research led to better understanding of problems unique to geographical areas with permafrost.