Third Generation Signal Integrity Tools and Issues

G. Katopis
{"title":"Third Generation Signal Integrity Tools and Issues","authors":"G. Katopis","doi":"10.1109/SPI.2002.258264","DOIUrl":null,"url":null,"abstract":"There exist two types of Signal integrity tool sets. One for the high level design of high performance packaging structures, and one for the verification of a package design after the interconnection wiring has been completed based on the rules and directions developed during the high level design. Not only the components of these tool sets are different, but also, even when the components of these tool sets are similar, their attributes are very different because of the different objectives that such components are required to fulfill. In this talk we will consider the tools used for the verification of the design of a package structure after its physical wiring has been completed. As we will show the ability to develop such tools presupposes the availability of adequate high level design signal integrity tools. This type of signal integrity tools includes accurate electrical parameter extractors for large physical structures, and accurate circuit analysis programs. It is to be noted that for this category of tools the main objective is their accuracy and ability to handle large physical structures, while their execution time is not as important for reasons that will become clear in the following. To the contrary for the verification tools the opposite is true. The development of both types of tools has started in IBM a long time ago, because of the needs of the mainframe computers that IBM has been provided to the market from the beginning of the computer age. Even the early versions of these systems required high frequency interconnections to support SMP servers ( Symmetric Multiprocessing ) that in turn required very large volumetric densities of the processor chips. These requirements resulted in the use of Multi Chip Modules (or MCM) namely, ceramic structures carrying large numbers of the Silicon chips ( either bipolar in the decade of 70s and 80s, or CMOS in the 90s ), and a large number of interconnections ranging from 5000 to 18000. In addition, the design of such complicated systems required the involvement of different skills from different engineering areas, and hence the need of comprehensive hand shakes for the progression of the package design from one phase to the next. Clearly, one of the most important phases is the delivery of the paper design of an MCM for its implementation to manufactured product. This is even more important for the ceramic MCM technology due to its long manufacturing TAT ( Turn Around Time). Therefore, the first generation of package design verification tools was developed in IBM in the 80s, in order to make sure that single pass design of MCMs could be achieved. The first generation of these tools was reflecting the technology needs and requirements but in addition, it established a set of attributes that such tools should have that transcends their particular application. In this presentation we shall describe these attributes that define the accuracy and execution time limitations, as well as the limitations that these tools had with respect to the interconnection topology considered. As the CMOS technology became the pervasive technology for all computing systems, and facilitated the significant increase of the interconnection speed and number of system components within one box, the needs of the mainframes in yesteryears became very similar to the needs for the field replaceable units of the current products.","PeriodicalId":290013,"journal":{"name":"Proceedings: 6th IEEE Workshop on Signal Propagation on Interconnects","volume":"160 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings: 6th IEEE Workshop on Signal Propagation on Interconnects","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SPI.2002.258264","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

There exist two types of Signal integrity tool sets. One for the high level design of high performance packaging structures, and one for the verification of a package design after the interconnection wiring has been completed based on the rules and directions developed during the high level design. Not only the components of these tool sets are different, but also, even when the components of these tool sets are similar, their attributes are very different because of the different objectives that such components are required to fulfill. In this talk we will consider the tools used for the verification of the design of a package structure after its physical wiring has been completed. As we will show the ability to develop such tools presupposes the availability of adequate high level design signal integrity tools. This type of signal integrity tools includes accurate electrical parameter extractors for large physical structures, and accurate circuit analysis programs. It is to be noted that for this category of tools the main objective is their accuracy and ability to handle large physical structures, while their execution time is not as important for reasons that will become clear in the following. To the contrary for the verification tools the opposite is true. The development of both types of tools has started in IBM a long time ago, because of the needs of the mainframe computers that IBM has been provided to the market from the beginning of the computer age. Even the early versions of these systems required high frequency interconnections to support SMP servers ( Symmetric Multiprocessing ) that in turn required very large volumetric densities of the processor chips. These requirements resulted in the use of Multi Chip Modules (or MCM) namely, ceramic structures carrying large numbers of the Silicon chips ( either bipolar in the decade of 70s and 80s, or CMOS in the 90s ), and a large number of interconnections ranging from 5000 to 18000. In addition, the design of such complicated systems required the involvement of different skills from different engineering areas, and hence the need of comprehensive hand shakes for the progression of the package design from one phase to the next. Clearly, one of the most important phases is the delivery of the paper design of an MCM for its implementation to manufactured product. This is even more important for the ceramic MCM technology due to its long manufacturing TAT ( Turn Around Time). Therefore, the first generation of package design verification tools was developed in IBM in the 80s, in order to make sure that single pass design of MCMs could be achieved. The first generation of these tools was reflecting the technology needs and requirements but in addition, it established a set of attributes that such tools should have that transcends their particular application. In this presentation we shall describe these attributes that define the accuracy and execution time limitations, as well as the limitations that these tools had with respect to the interconnection topology considered. As the CMOS technology became the pervasive technology for all computing systems, and facilitated the significant increase of the interconnection speed and number of system components within one box, the needs of the mainframes in yesteryears became very similar to the needs for the field replaceable units of the current products.
第三代信号完整性工具和问题
存在两种类型的信号完整性工具集。一个用于高性能封装结构的高级设计,另一个用于根据高级设计期间制定的规则和方向,在互连布线完成后对封装设计进行验证。不仅这些工具集的组件是不同的,而且,即使这些工具集的组件是相似的,它们的属性也非常不同,因为这些组件需要实现不同的目标。在这次演讲中,我们将考虑在物理布线完成后用于验证封装结构设计的工具。正如我们将展示的那样,开发此类工具的能力以足够的高层次设计信号完整性工具的可用性为前提。这种类型的信号完整性工具包括用于大型物理结构的精确电参数提取器和精确的电路分析程序。值得注意的是,对于这类工具,主要目标是它们的准确性和处理大型物理结构的能力,而它们的执行时间并不重要,原因将在下面变得清楚。与验证工具相反,情况正好相反。这两种工具的开发很早就在IBM开始了,因为IBM从计算机时代开始就向市场提供了大型计算机的需求。即使是这些系统的早期版本也需要高频互连来支持SMP服务器(对称多处理),这反过来又需要非常大的处理器芯片体积密度。这些要求导致使用多芯片模块(或MCM),即携带大量硅芯片的陶瓷结构(70年代和80年代的双极,或90年代的CMOS),以及从5000到18000的大量互连。此外,这种复杂系统的设计需要来自不同工程领域的不同技能的参与,因此需要从一个阶段到下一个阶段的包装设计进展的全面握手。显然,最重要的阶段之一是将MCM的纸面设计交付到制造产品中。这对于陶瓷MCM技术来说更为重要,因为它的制造TAT(周转时间)很长。因此,第一代封装设计验证工具在80年代由IBM开发出来,以确保mcm的单通道设计能够实现。这些工具的第一代反映了技术需求和需求,但除此之外,它还建立了这些工具应该具有的一组属性,这些属性超出了它们的特定应用。在本演示中,我们将描述这些定义准确性和执行时间限制的属性,以及这些工具相对于所考虑的互连拓扑的限制。随着CMOS技术成为所有计算系统的普遍技术,并促进了一个盒子内系统组件互连速度和数量的显着增加,过去对大型机的需求与当前产品对现场可更换单元的需求非常相似。
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
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信