CAREER: Practical Privacy for Outsourcing Systems

该奖项是根据2009年美国复苏和再投资法案（公法111-5）资助的。随着计算嵌入到我们社会的结构中，廉价、高速通信基础设施的指数级增长和进步使全球信息交换和交互达到了前所未有的水平。因此，新的市场力量出现了，推动了计算部署和交付方式的根本性、成本效益的范式转变：计算外包。外包有可能最大限度地减少客户端管理开销，并从服务提供商的全球专业知识整合和批量定价中获益。谷歌、雅虎、亚马逊和Sun等公司都急于提供由全球分布式“云”基础设施支持的日益复杂的存储和计算外包服务。然而，在成功大规模采用的道路上存在着重大挑战。在商业、医疗保健和政府框架中，客户不愿意将敏感数据置于远程第三方提供商的控制之下，而没有实际的隐私和机密性保证。然而，今天的解决方案并不提供这样的保证，因此从根本上讲是不安全的，容易受到非法行为的影响。现有的研究解决了这个问题的几个方面，但将技术水平提升到实用领域需要根本性的飞跃。该项目通过设计、实施和分析具有强有力的隐私和机密性保证的实用数据外包协议来解决这些挑战。 它还将启动对外包机制的成本和能源足迹的探索。这一点至关重要，因为外包模式的主要存在理由在于其假定的端到端成本节约和专业知识整合。然而，研究尚未探索和验证这些节省的幅度及其基本假设。

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). As computing becomes embedded in the very fabric of our society, theexponential growth and advances in cheap, high-speed communicationinfrastructures allow for unprecedented levels of global informationexchange and interaction. As a result, new market forces emerge that propeltoward a fundamental, cost-efficient paradigm shift in the way computing isdeployed and delivered: computing outsourcing.Outsourcing has the potential to minimize client-side management overheadsand benefit from a service provider's global expertise consolidation andbulk pricing. Companies such as Google, Yahoo, Amazon, and Sun are rushingto offer increasingly complex storage and computation outsourcing servicessupported by globally distributed "cloud" infrastructures.Yet significant challenges lie in the path to successful large-scaleadoption. In business, healthcare and government frameworks, clients arereluctant to place sensitive data under the control of a remote, third-party provider, without practical assurances of privacy and confidentiality.Today's solutions however, do not offer such assurances, and are thusfundamentally insecure and vulnerable to illicit behavior. Existing researchaddresses several aspects of this problem, but advancing the state of theart to practical realms will require a fundamental leap.This project addresses these challenges by designing, implementing andanalyzing practical data outsourcing protocols with strong assurances ofprivacy and confidentiality. It will also l initiate the exploration of the cost and energyfootprints of outsourcing mechanisms. This is essential as the main raisond'etre of outsourcing paradigms lies in their assumed end-to-end costsavings and expertise consolidation. Yet, research has yet to explore andvalidate the magnitudes of these savings and their underlying assumptions.