{"title":"ASCI Pathforward: to 30 Tflops and beyond","authors":"D. Clark","doi":"10.1109/4434.678783","DOIUrl":null,"url":null,"abstract":"Future computer historians may speak of the technology advances developed for the US Department of Energy's Accelerated Strategic Computing Initiative the way today's historians speak about technology that came out of DARPA in the 1970s and 80s, with awe. DARPA projects produced innovation that led to the Internet and Unix, among other things. ASCI has already produced the first teraflops computer. If the confidence of companies who recently signed contracts for ASCI's Pathforward project is any indication, we will see a computer capable of 30 Tflops in 2001. However, because of ASCI's business model, which requires commercial off-the-shelf components, it is unlikely that ASCI will get credit for much of the technology. The author discusses two plans for ASCI to leverage existing research and development. He also considers mass storage and software issues.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1998-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Concurr.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/4434.678783","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Future computer historians may speak of the technology advances developed for the US Department of Energy's Accelerated Strategic Computing Initiative the way today's historians speak about technology that came out of DARPA in the 1970s and 80s, with awe. DARPA projects produced innovation that led to the Internet and Unix, among other things. ASCI has already produced the first teraflops computer. If the confidence of companies who recently signed contracts for ASCI's Pathforward project is any indication, we will see a computer capable of 30 Tflops in 2001. However, because of ASCI's business model, which requires commercial off-the-shelf components, it is unlikely that ASCI will get credit for much of the technology. The author discusses two plans for ASCI to leverage existing research and development. He also considers mass storage and software issues.