CSR: DMSS, TM: Quantifying the Impact of Latency on Network-Interactive Computing

在交互式计算模型中，位于用户交互关键路径上的代码段跨越网络边界被分割，这引起了学术界和商界越来越多的兴趣。我们使用术语“网络交互计算”来描述这种模型。瘦客户机计算和基于ajax的计算就是两个例子。不幸的是，在用户交互的关键路径中引入网络延迟会降低系统响应的灵敏性，并损害可用性。今天，我们缺乏科学的框架来量化延迟对网络交互计算的影响。这项研究将在性能和可用性方面建立网络交互计算的科学基础。我们的工作将涉及三个领域：(a)制定指标和基准；(b)根据用户经验校准指标；(c)根据用户交互特征对实际应用进行分类。我们将开发智能工具和评估技术来帮助回答以下问题：“在本地计算机上运行良好的遗留交互式应用程序在特定的网络交互式计算环境中是否会不可接受地降级？”“当设计一个网络交互计算的新环境时，我们能预测哪些应用程序的性能是可以接受的吗？”以及“是否可以开发应用程序的分类，使我们能够轻松地理解和预测网络变化对交互性能的影响？”“我们的目标是在一个健全的、经过充分验证的、可复制的实验基础上建立网络交互计算。

There is growing academic and commercial interest in models ofinteractive computing where code segments that lie in the criticalpath of user interaction are split across a network boundary. We usethe term "Network-Interactive Computing" to characterize such models.Thin client computing and Ajax-based computing are twoexamples. Unfortunately, introducing network delays into the criticalpath of user interaction can degrade the crispness of system response,and hurt usability. Today, we lack the scientific framework toquantify the impact of latency on network-interactive computing.This research will establish the scientific underpinnings ofnetwork-interactive computing with respect to performance andusability. Our work will span three areas: (a) Development of metricsand benchmarks; (b) Calibration of metrics against user experience;and (c) Classification of real-world applications based on their userinteraction characteristics. We will develop the intellectual toolsand evaluation techniques to help answer questions such as thefollowing: ``Will a legacy interactive application that performs wellon a local computer degrade unacceptably in a specifiednetwork-interactive computing environment?''; ``When designing a newenvironment for network-interactive computing, can one predict whichapplications will perform acceptably?''; and, ``Can one develop ataxonomy of applications that allows us to easily understand andpredict the impact of networking changes on interactive performance?''Our goal is to establish network-interactive computing on a sound,well-validated and replicatable experimental foundation.