Distributed Execution with Remote Audit

By Fabian Monrose, P. Wyckoff, Aviel D. Rubin. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Verifying computations with state (2013, background): “Indeed, when a job completes, after having processed petabytes of data, how can the client be sure that the output is correct [77]? The client could audit the output [55], but this technique fails if a problem happens outside the selected sample.”
SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloud (2017, unknown): “Audits based on the server’s execution time [14] require precise knowledge of the server’s hardware conﬁguration and assume, for instance, that the server is not over-clocked.”
Efficient RAM and control flow in verifiable outsourced computation (2015, background): “Buffet allows the programmer to express programs in an expansive subset of C (disallowing only “goto” and function pointers), can handle essentially any example in the verifiable computation literature, and achieves the best performance in the area by multiple orders of magnitude.”
Making argument systems for outsourced computation practical (sometimes) (2012, background): “Auditing and spot-checking [31, 42, 46, 50, 57] focus directly on the output, but unless the audit coverage is nearly perfect (which degenerates to executing the computation), a worker that alters a key bit (e.”
Resolving the conflict between generality and plausibility in verified computation (2013, background): “Auditing [43] can be effective for highly structured computations but assumes detailed knowledge of the intermediate results.”
Secure Distributed Computing in a Commercial Environment (2002, methodology): “See for example [ MWR99 ] for a very general technique.”
Security and privacy for storage and computation in cloud computing (2014, methodology): “[27] presented a remote audit mechanism on an existing distributed computing model and provided efficient methods for verifying whether a remote host performed the assigned task.”
Obfuscator-LLVM -- Software Protection for the Masses (2015, background): “So far, we observe that the problem of protecting the integrity and con.dentiality of running code has been attacked from two opposite directions, namely cryptographic and software obfuscation.”
Searching for High-Value Rare Events with Uncheatable Grid Computing (2005, background): “However the tamper-resistant approach is only heuristically secure, and many tamper-resistant schemes cannot withstand attacks from coalitions of determined attackers [11].”
Publicly verifiable delegation of large polynomials and matrix computations, with applications (2012, background): “In the past, more practical solutions, but of limited provable security, were also proposed: e.g., solutions based on audit (e.g. [19, 3]) or secure co-processors (e.g. [22, 23]) which ”sign” the computation as correct, under the assumption that the adversary cannot tamper with the processor.”
A Hybrid Architecture for Interactive Verifiable Computation (2013, unknown): “Experimental results indicate that our system performs better and applies more widely than the best in the literature.”
Verifiable Delegation of Computation over Large Datasets (2011, background): “[46,6]) which typically assume many clients, and require a fraction of them to recompute some of the results provided by the server, but are secure only under the assumption that bad clients do not collude.”
Non-interactive Verifiable Computing: Outsourcing Computation to Untrusted Workers (2010, background): “The public information is sent to the worker.”