CAREER: Practical verification of outsourced computations

点击翻译按钮获取中文摘要

Many computing services are moving to the "cloud" because cloudcomputing offers economic and operational benefits. However, cloudservices not only are complex black boxes operated by third parties butalso refuse to give service guarantees. Thus, instead of trying to makecloud services more trustworthy (that is, giving customers reasons toplace faith in the cloud), this project explicitly removes trust fromthe cloud: the project is building systems that give cloud customersguarantees, even under arbitrary service malfunctions. Two such systemsconcern storage and function computation.Under the storage system, the challenge is to use client-sideverification to resist arbitrary or malicious cloud malfunctions (losingdata, returning stale data, corrupting data, compromising data, etc.).Under the function computation system, a client specifies a computationto a server, the server executes it and returns the output and auxiliaryinformation, and the client uses the auxiliary information toefficiently check whether the output is correct. The approach is toadapt, and translate into practice, a fascinating body of theory (onprobabilistically checkable and interactive proofs) that is currentlythought to be impractical.With qualitative advances in reliable yet practical computing, thisproject will make cloud computing safer for existing customers and spurthe adoption of cloud services. This will mean more people paying lessfor computing, producing beneficial effects throughout the computationalecosystem. The educational thrusts include teaching graduate students toteach (applying the K-12 student teaching model), graduate andundergraduate mentoring, curriculum development, and outreach tounderrepresented groups.