Biblio

A number of small Open Source projects let independent providers measure different aspects of their quality that would otherwise be hard to see. This paper describes this observation as the pattern Quality Attestation. Quality Attestation belongs to a family of Open Source patterns written by various authors.

Web applications are a frequent target of successful attacks. In most web frameworks, the damage is amplified by the fact that application code is responsible for security enforcement. In this paper, we design and evaluate Radiatus, a shared-nothing web framework where application-specific computation and storage on the server is contained within a sandbox with the privileges of the end-user. By strongly isolating users, user data and service availability can be protected from application vulnerabilities. To make Radiatus practical at the scale of modern web applications, we introduce a distributed capabilities system to allow fine-grained secure resource sharing across the many distributed services that compose an application. We analyze the strengths and weaknesses of a shared-nothing web architecture, which protects applications from a large class of vulnerabilities, but adds an overhead of 60.7% per server and requires an additional 31MB of memory per active user. We demonstrate that the system can scale to 20K operations per second on a 500-node AWS cluster.

Let G be a finite connected graph, and let ρ be the spectral radius of its universal cover. For example, if G is k-regular then ρ=2√k−1. We show that for every r, there is an r-covering (a.k.a. an r-lift) of G where all the new eigenvalues are bounded from above by ρ. It follows that a bipartite Ramanujan graph has a Ramanujan r-covering for every r. This generalizes the r=2 case due to Marcus, Spielman and Srivastava (2013). Every r-covering of G corresponds to a labeling of the edges of G by elements of the symmetric group Sr. We generalize this notion to labeling the edges by elements of various groups and present a broader scenario where Ramanujan coverings are guaranteed to exist. In particular, this shows the existence of richer families of bipartite Ramanujan graphs than was known before. Inspired by Marcus-Spielman-Srivastava, a crucial component of our proof is the existence of interlacing families of polynomials for complex reflection groups. The core argument of this component is taken from Marcus-Spielman-Srivastava (2015). Another important ingredient of our proof is a new generalization of the matching polynomial of a graph. We define the r-th matching polynomial of G to be the average matching polynomial of all r-coverings of G. We show this polynomial shares many properties with the original matching polynomial. For example, it is real rooted with all its roots inside [−ρ,ρ].

Many emerging applications, from domains such as healthcare and oil & gas, require several data processing systems for complex analytics. This demo paper showcases system, a framework that provides multi-platform task execution for such applications. It features a three-layer data processing abstraction and a new query optimization approach for multi-platform settings. We will demonstrate the strengths of system by using real-world scenarios from three different applications, namely, machine learning, data cleaning, and data fusion.

Consensus algorithms provide strategies to solve problems in a distributed system with the added constraint that data can only be shared between adjacent computing nodes. We find these algorithms in applications for wireless and sensor networks, spectrum sensing for cognitive radio, even for some IoT services. However, consensus-based applications are not resilient to compromised nodes sending falsified data to their neighbors, i.e. they can be the target of Byzantine attacks. Several solutions have been proposed in the literature inspired from reputation based systems, outlier detection or model-based fault detection techniques in process control. We have reviewed some of these solutions, and propose two mitigation techniques to protect the consensus-based Network Intrusion Detection System in [1]. We analyze several implementation issues such as computational overhead, fine tuning of the solution parameters, impacts on the convergence of the consensus phase, accuracy of the intrusion detection system.

Consensus algorithms provide strategies to solve problems in a distributed system with the added constraint that data can only be shared between adjacent computing nodes. We find these algorithms in applications for wireless and sensor networks, spectrum sensing for cognitive radio, even for some IoT services. However, consensus-based applications are not resilient to compromised nodes sending falsified data to their neighbors, i.e. they can be the target of Byzantine attacks. Several solutions have been proposed in the literature inspired from reputation based systems, outlier detection or model-based fault detection techniques in process control. We have reviewed some of these solutions, and propose two mitigation techniques to protect the consensus-based Network Intrusion Detection System in [1]. We analyze several implementation issues such as computational overhead, fine tuning of the solution parameters, impacts on the convergence of the consensus phase, accuracy of the intrusion detection system.

The evolution of the Internet of Things (IoT) requires a well-defined infrastructure of systems that provides services for device abstraction and data management, and also supports the development of applications. Middleware for IoT has been recognized as the system that can provide these services and has become increasingly important for IoT in recent years. The large amount of data that flows into a middleware system demands a security architecture that ensures the protection of all layers of the system, including the communication channels and border APIs used to integrate the applications and IoT devices. However, this security architecture should be based on lightweight approaches since middleware systems are widely applied in constrained environments. Some works have already defined new solutions and adaptations to existing approaches in order to mitigate IoT middleware security problems. In this sense, this article discusses the role of lightweight approaches to the standardization of a security architecture for IoT middleware systems. This article also analyzes concepts and existing works, and presents some important IoT middleware challenges that may be addressed by emerging lightweight security approaches in order to achieve the consolidation of a standard security architecture and the mitigation of the security problems found in IoT middleware systems.

As web applications is becoming more prominent due to the ubiquity of web services, web applications have become main targets for attackers. In order to steal or leak sensitive user data managed by web applications, attackers exploit a wide range of input validation vulnerabilities such as SQL injection, path traversal (or directory traversal), cross-site scripting (XSS), etc. This paper propose a technique that can verify input values of Java-based web applications using static bytecode instrumentation and runtime input validation. The technique searches for target methods or object constructors in compiled Java class files, and statically inserts bytecode modules. At runtime, the instrumented bytecode modules validate input values of the targets, and take countermeasure against malicious inputs. The proposed technique can mitigate the input validation vulnerabilities in Java-based web applications without source codes. To evaluate the effectiveness of the proposed technique, experiments are carried out with an insecure web application maintained by OWASP WebGoat Project. The experimental results show that the proposed technique successfully mitigates input validation vulnerabilities such as SQL injection and path traversal.

Users of modern data-processing services such as tax preparation or genomic screening are forced to trust them with data that the users wish to keep secret. Ryoan protects secret data while it is processed by services that the data owner does not trust. Accomplishing this goal in a distributed setting is difficult because the user has no control over the service providers or the computational platform. Confining code to prevent it from leaking secrets is notoriously difficult, but Ryoan benefits from new hardware and a request-oriented data model. Ryoan provides a distributed sandbox, leveraging hardware enclaves (e.g., Intel's software guard extensions (SGX) [15]) to protect sandbox instances from potentially malicious computing platforms. The protected sandbox instances confine untrusted data-processing modules to prevent leakage of the user's input data. Ryoan is designed for a request-oriented data model, where confined modules only process input once and do not persist state about the input. We present the design and prototype implementation of Ryoan and evaluate it on a series of challenging problems including email filtering, heath analysis, image processing and machine translation.

It is a fundamental problem to decide how many copies of an unknown mixed quantum state are necessary and sufficient to determine the state. This is the quantum analogue of the problem of estimating a probability distribution given some number of samples. Previously, it was known only that estimating states to error є in trace distance required O(dr2/є2) copies for a d-dimensional density matrix of rank r. Here, we give a measurement scheme (POVM) that uses O( (dr/ δ ) ln(d/δ) ) copies to estimate ρ to error δ in infidelity. This implies O( (dr / є2)· ln(d/є) ) copies suffice to achieve error є in trace distance. For fixed d, our measurement can be implemented on a quantum computer in time polynomial in n. We also use the Holevo bound from quantum information theory to prove a lower bound of Ω(dr/є2)/ log(d/rє) copies needed to achieve error є in trace distance. This implies a lower bound Ω(dr/δ)/log(d/rδ) for the estimation error δ in infidelity. These match our upper bounds up to log factors. Our techniques can also show an Ω(r2d/δ) lower bound for measurement strategies in which each copy is measured individually and then the outcomes are classically post-processed to produce an estimate. This matches the known achievability results and proves for the first time that such “product” measurements have asymptotically suboptimal scaling with d and r.

It is a fundamental problem to decide how many copies of an unknown mixed quantum state are necessary and sufficient to determine the state. This is the quantum analogue of the problem of estimating a probability distribution given some number of samples. Previously, it was known only that estimating states to error є in trace distance required O(dr2/є2) copies for a d-dimensional density matrix of rank r. Here, we give a measurement scheme (POVM) that uses O( (dr/ δ ) ln(d/δ) ) copies to estimate ρ to error δ in infidelity. This implies O( (dr / є2)· ln(d/є) ) copies suffice to achieve error є in trace distance. For fixed d, our measurement can be implemented on a quantum computer in time polynomial in n. We also use the Holevo bound from quantum information theory to prove a lower bound of Ω(dr/є2)/ log(d/rє) copies needed to achieve error є in trace distance. This implies a lower bound Ω(dr/δ)/log(d/rδ) for the estimation error δ in infidelity. These match our upper bounds up to log factors. Our techniques can also show an Ω(r2d/δ) lower bound for measurement strategies in which each copy is measured individually and then the outcomes are classically post-processed to produce an estimate. This matches the known achievability results and proves for the first time that such “product” measurements have asymptotically suboptimal scaling with d and r.

Attack graphs are a powerful modeling technique with which to explore the attack surface of a system. However, they can be difficult to generate due to the exponential growth of the state space, often times making exhaustive search impractical. This paper discusses an approach for generating large attack graphs with an emphasis on scalable generation over a distributed system. First, a serial algorithm is presented, highlighting bottlenecks and opportunities to exploit inherent concurrency in the generation process. Then a strategy to parallelize this process is presented. Finally, we discuss plans for future work to implement the parallel algorithm using a hybrid distributed/shared memory programming model on a heterogeneous compute node cluster.

With the rapid growth of network bandwidth, increases in CPU cores on a single machine, and application API models demanding more short-lived connections, a scalable TCP stack is performance-critical. Although many clean-state designs have been proposed, production environments still call for a bottom-up parallel TCP stack design that is backward-compatible with existing applications. We present Fastsocket, a BSD Socket-compatible and scalable kernel socket design, which achieves table-level connection partition in TCP stack and guarantees connection locality for both passive and active connections. Fastsocket architecture is a ground up partition design, from NIC interrupts all the way up to applications, which naturally eliminates various lock contentions in the entire stack. Moreover, Fastsocket maintains the full functionality of the kernel TCP stack and BSD-socket-compatible API, and thus applications need no modifications. Our evaluations show that Fastsocket achieves a speedup of 20.4x on a 24-core machine under a workload of short-lived connections, outperforming the state-of-the-art Linux kernel TCP implementations. When scaling up to 24 CPU cores, Fastsocket increases the throughput of Nginx and HAProxy by 267% and 621% respectively compared with the base Linux kernel. We also demonstrate that Fastsocket can achieve scalability and preserve BSD socket API at the same time. Fastsocket is already deployed in the production environment of Sina WeiBo, serving 50 million daily active users and billions of requests per day.

We propose the first verifier-local revocation scheme for privacy-enhancing attribute-based credentials (PABCs) that is practically usable in large-scale applications, such as national eID cards, public transportation and physical access control systems. By using our revocation scheme together with existing PABCs, it is possible to prove attribute ownership in constant time and verify the proof and the revocation status in the time logarithmic in the number of revoked users, independently of the number of all valid users in the system. Proofs can be efficiently generated using only offline constrained devices, such as existing smart-cards. These features are achieved by using a new construction called \$n\$-times unlinkable proofs. We show the full cryptographic description of the scheme, prove its security, discuss parameters influencing scalability and provide details on implementation aspects. As a side result of independent interest, we design a more efficient proof of knowledge of weak Boneh-Boyen signatures, that does not require any pairing computation on the prover side.

Software-Defined Networking (SDN) has emerged as a framework for centralized command and control in cloud data centric environments. SDN separates data and control plane, which provides network administrator better visibility and policy enforcement capability compared to traditional networks. The SDN controller can assess reachability information of all the hosts in a network. There are many critical assets in a network which can be compromised by a malicious attacker through a multistage attack. Thus we make use of centralized controller to assess the security state of the entire network and pro-actively perform attack analysis and countermeasure selection. This approach is also known as Moving Target Defense (MTD). We use the SDN controller to assess the attack scenarios through scalable Attack Graphs (AG) and select necessary countermeasures to perform network reconfiguration to counter network attacks. Moreover, our framework has a comprehensive conflict detection and resolution module that ensures that no two flow rules in a distributed SDN-based cloud environment have conflicts at any layer; thereby assuring consistent conflict-free policy implementation and preventing information leakage.

Software Defined Networking (SDN) is the new promise towards an easily configured and remotely controlled network. Based on Centralized control, SDN technology has proved its positive impact on the world of network communications from different aspects. Security in SDN, as in traditional networks, is an essential feature that every communication system should possess. In this paper, we propose an SDN security design approach, which strikes a good balance between network performance and security features. We show how such an approach can be used to prevent DDoS attacks targeting either the controller or the different hosts in the network, and how to trace back the source of the attack. The solution lies in introducing a third plane, the security plane, in addition to the data plane, which is responsible for forwarding data packets between SDN switches, and parallel to the control plane, which is responsible for rule and data exchange between the switches and the SDN controller. The security plane is designed to exchange security-related data between a third party agent on the switch and a third party software module alongside the controller. Our evaluation shows the capability of the proposed system to enforce different levels of real-time user-defined security with low overhead and minimal configuration.