Throughput computing

Proceedings. 11th IEEE International Conference and Workshop on the Engineering of Computer-Based Systems, 2004. Pub Date : 1900-01-01 DOI:10.1109/ECBS.2004.1316733

Petr Matena

{"title":"Throughput computing","authors":"Petr Matena","doi":"10.1109/ECBS.2004.1316733","DOIUrl":null,"url":null,"abstract":"Summary form only given. Customers today spend 70% of their IT budget on current operations with only 30% going towards adding new services and gaining competitive advantage. They would rather have the proportions reversed. Sun's throughput computing strategy aims to help our customers get there. The significantly increased processor throughput enable customers do a whole lot more with less. For a given workload, the IT infrastructure costs such as floor space, power, and maintenance is dramatically reduced. Additionally, fewer components per system also increase the reliability of the systems. Basically the cost of computing is reduced by an order of magnitude. This is the biggest change in computing technology since we moved from tubes to transistors to integrated microprocessors. And it frees customers to spend a much bigger proportion of their IT budget on deploying entirely new and more competitive services via network computing. Executing on Sun's throughput computing strategy, the dual-threaded UltraSPARC IV processor marks the first milestone in the CMT processor roadmap from Sun.","PeriodicalId":137219,"journal":{"name":"Proceedings. 11th IEEE International Conference and Workshop on the Engineering of Computer-Based Systems, 2004.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. 11th IEEE International Conference and Workshop on the Engineering of Computer-Based Systems, 2004.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ECBS.2004.1316733","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Summary form only given. Customers today spend 70% of their IT budget on current operations with only 30% going towards adding new services and gaining competitive advantage. They would rather have the proportions reversed. Sun's throughput computing strategy aims to help our customers get there. The significantly increased processor throughput enable customers do a whole lot more with less. For a given workload, the IT infrastructure costs such as floor space, power, and maintenance is dramatically reduced. Additionally, fewer components per system also increase the reliability of the systems. Basically the cost of computing is reduced by an order of magnitude. This is the biggest change in computing technology since we moved from tubes to transistors to integrated microprocessors. And it frees customers to spend a much bigger proportion of their IT budget on deploying entirely new and more competitive services via network computing. Executing on Sun's throughput computing strategy, the dual-threaded UltraSPARC IV processor marks the first milestone in the CMT processor roadmap from Sun.

查看原文本刊更多论文

吞吐量计算

只提供摘要形式。如今，客户将70%的IT预算用于当前运营，只有30%用于增加新服务和获得竞争优势。他们宁愿比例颠倒过来。Sun的吞吐量计算策略旨在帮助我们的客户实现这一目标。显著提高的处理器吞吐量使客户能够事半功倍。对于给定的工作负载，IT基础设施成本(如占地面积、电力和维护)大大降低。此外，每个系统较少的组件也增加了系统的可靠性。基本上，计算成本降低了一个数量级。这是自从我们从电子管到晶体管再到集成微处理器以来，计算技术的最大变化。而且，它让客户可以将更多的it预算用于通过网络计算部署全新的、更具竞争力的服务。执行Sun的吞吐量计算策略，双线程UltraSPARC IV处理器标志着Sun CMT处理器路线图的第一个里程碑。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Proceedings. 11th IEEE International Conference and Workshop on the Engineering of Computer-Based Systems, 2004.

自引率

0.00%

发文量