More and more users store data in "clouds" that are accessed remotely over the Internet. We survey well-known cryptographic tools for providing integrity consistency stored clouds discuss recent research cryptography distributed computing addressing these problems.

PU hardware is becoming increasingly general purpose, quickly outgrowing the traditional but constrained GPU-as-coprocessor programming model. To make GPUs easier to program and integrate with existing systems, we propose making host's file system directly accessible from GPU code. GPUfs provides a POSIX-like API for programs, exploits parallelism efficiency, optimizes access by extending buffer cache into memory. Our experiments, based on set of real benchmarks adopted use our system,...

We consider the problem of in-network compressed sensing from distributed measurements. Every agent has a set measurements signal x, and objective is for agents to recover x their collective using only communication with neighbors in network. Our approach this based on centralized Iterative Hard Thresholding algorithm (IHT). first present IHT static networks that leverages standard tools computing execute computations minimized bandwidth consumption. Next, we address recovery time-varying...

We present DAG-Rider, the first asynchronous Byzantine Atomic Broadcast protocol that achieves optimal resilience, amortized communication complexity, and time complexity. DAG-Rider is post-quantum safe ensures all values proposed by correct processes eventually get delivered. construct in two layers: In layer, reliably broadcast their proposals build a structured Directed Acyclic Graph (DAG) of among them. second locally observe DAGs totally order with no extra communication.

This paper presents Venus, a service for securing user interaction with untrusted cloud storage. Specifically, Venus guarantees integrity and consistency applications accessing key-based object store service, without requiring trusted components or changes to the storage provider. completes all operations optimistically, guaranteeing data integrity. It then verifies operation notifies application. Whenever either is violated, alerts We implemented evaluated it Amazon S3 commodity service....

An effective way to reduce the number of aborts in software transactional memory (STM) is keep multiple versions objects. In this paper, we study inherent properties STMs that use guarantee successful commits all read-only transactions.

We study the tradeoffs between many-core machines like Intelpsilas Larrabee and many-thread Nvidia AMD GPGPUs. define a unified model describing superposition of two architectures, use it to identify operation zones for which each machine is more suitable. Moreover, we an intermediate zone in both deliver inferior performance. shape this ldquoperformance valleyrdquo provide insights on how can be avoided.

This article deals with the emulation of atomic read/write (R/W) storage in dynamic asynchronous message passing systems. In static settings, it is well known that R/W can be implemented a fault-tolerant manner even if system completely asynchronous, whereas consensus not solvable. contrast, all existing emulations systems rely on or stronger primitives, leading to popular belief unattainable without consensus. this article, we specify problem terms interface available users such storage. We...

Log-structured data stores (LSM-DSs) are widely accepted as the state-of-the-art implementation of key-value stores. They replace random disk writes with sequential I/O, by accumulating large batches updates in an in-memory structure and merging it on-disk store background. While LSM-DS implementations proved to be highly successful at masking I/O bottleneck, scaling them up on multicore CPUs remains a challenge. This is nontrivial due their often rich APIs, well need coordinate RAM access...

As GPU hardware becomes increasingly general-purpose, it is quickly outgrowing the traditional, constrained GPU-as-coprocessor programming model. This article advocates for extending standard operating system services and abstractions to GPUs in order facilitate program development enable harmonious integration of computing systems. an example, we describe design implementation GPUFs, a software layer which provides support accessing host files directly from programs. GPUFs POSIX-like API,...

We provide a suite of impossibility results and lower bounds for the required number processes rounds synchronous consensus under transient link failures. Our show that can be solved even in presence $O(n^2)$ moving omission and/or arbitrary failures per round, provided both affected outgoing incoming links every process is bounded. Providing step further toward weakest conditions which solvable, our findings are applicable to variety dynamic phenomena such as communication end-to-end delay...

No abstract available.

We consider the problem of overcoming (distributed) denial-of-service (DoS) attacks by realistic adversaries that have knowledge their attack's successfulness, for example, observing service performance degradation or eavesdropping on messages parts thereof. A solution this in a high-speed network environment necessitates lightweight mechanisms differentiating between valid traffic and attacker's packets. The main challenge presenting such is to exploit existing packet-filtering way allows...

We present a novel network-on-chip-based architecture for future programmable chips (FPGAs). A key challenge FPGA design is supporting numerous highly variable instances with good performance and low cost. Our minimizes the cost of wide range given throughput requirements by balancing amount efficient hard-coded NoC infrastructure allocation "soft" networking resources at configuration time. Although traffic patterns are design-specific, physical link bottleneck, hence should be hard-coded....

We introduce transactions into libraries of concurrent data structures; such can be used to ensure atomicity sequences structure operations. By focusing on transactional access a well-defined set operations, we strike balance between the ease-of-programming and efficiency custom-tailored structures. exemplify this concept by designing implementing library supporting any number maps, sets (implemented as skiplists), queues. Our offers efficient scalable transactions, which are an order...

Log-structured merge (LSM) stores have emerged as the technology of choice for building scalable write-intensive key-value storage systems. An LSM store replaces random I/O with sequential by accumulating large batches writes in a memory prior to flushing them log-structured disk storage; latter is continuously re-organized background through compaction process efficiency reads. Though inherent design, frequent compactions are major pain point because they slow down data operations,...

We present Moshe, a novel scalable group membership algorithm built specifically for use in wide area networks (WANs), which can suffer partitions. Moshe is designed with three new significant features that are important this setting: it avoids delivering views reflect out-of-date memberships; requires single round of messages the common case; and employs client-server design scalability. Furthermore, Moshe's interface supplies hooks needed to provide clients full virtual synchrony...

We consider the consensus problem in an asynchronous model enriched with unreliable failure detectors and partial synchrony model. algorithms that solve tolerate crash failures and/or message omissions. prove tight lower bounds on number of communication steps performed by such failure-free executions. present a unified framework related bound results. Thus, we shed light relationships among different known upper bounds, at same time, illustrate general technique for obtaining simple elegant...

Article Free Access Share on Efficient message ordering in dynamic networks Authors: Idit Keidar Computer Science Institute, The Hebrew University of Jerusalem JerusalemView Profile , Danny Dolev Authors Info & Claims PODC '96: Proceedings the fifteenth annual ACM symposium Principles distributed computingMay 1996 Pages 68–76https://doi.org/10.1145/248052.248062Online:01 May 1996Publication History 40citation364DownloadsMetricsTotal Citations40Total Downloads364Last 12 Months17Last 6 weeks6...

We consider the problem of dust: Different URLs with Similar Text. Such duplicate are prevalent in web sites, as server software often uses aliases and redirections, translates to some canonical form, dynamically generates same page from various different URL requests. present a novel algorithm, DustBuster, for uncovering dust; that is, discovering rules transforming given others likely have similar content. DustBuster is able detect dust effectively previous crawl logs or logs, without...