We design and implement a scalable version of loopy belief propagation (BP), a widely used algorithm for performing inference on probabilistic graphical models. However, implementations of BP on generic data processing platforms such as Apache Spark do not scale well to very large graphical models containing billions of vertices. To handle such large-scale graphs, we leverage a number of strategies. Our implementation is based on Apache Spark GraphX. We propose a novel graph partitioning strategy to reduce both computation and communication overhead providing a 2x speed-up. We use efficient memory management for storing the graph and shared memory for highspeed communication. To evaluate performance and demonstrate scalability of the approach, we perform a range of experiments including using real-world graphs with billions of vertices, where we achieve an overall 10x speed-up over a vanilla Spark baseline. Further, we apply our BP implementation to infer the probability of a website being malicious by performing inference on a graphical model derived from real, large-scale hyperlinked webcrawl data. We have open sourced our implementation.

Given the growing availability of affordable scale-up servers, our goal is to bring the performance benefits of in-memory processing on scale-up servers to an increasingly common class of data analytics applications that process small to medium size datasets (up to a few 100GBs) that can easily fit in the memory of a typical scale-up server To achieve this goal, we leverage Spark, an existing memory-centric data analytics framework with wide-spread adoption among data scientists. Bringing Spark's data analytic capabilities to a scale-up system requires rethinking the original design assumptions, which, although effective for a scale-out system, are a poor match to a scale-up system resulting in unnecessary communication and memory inefficiencies.

Many modern large graph and Big Data processing applications operate on datasets that do not fit into DRAM of a single machine. This leads to a design of scale-out applications, where the application dataset is partitioned and processed by a cluster of machines. Distributed memory applications exhibit complex behavior: they tend to interleave computations and communications, use bursty transfers, and utilize global synchronization primitives. This makes it difficult to analyze the impact of communication layer on the application performance and answer the questions: how interconnect latency or bandwidth characteristics may change the application performance will the application performance scale when processed by a larger system? In this work, we introduce a novel emulation framework, called InterSense, which is implemented on top of existing high-speed interconnect, such as InfiniBand, and which provides two performance knobs for changing the (today's) interconnect bandwidth and latency. This approach offers an easy-to-use framework for a sensitivity analysis of complex distributed applications to communication layer performance instead of creating customized and time-consuming application models to answer the same questions. We evaluate the emulator accuracy with popular OSU MPI benchmark suite and two clusters with different generation InfiniBand interconnects (DDR and FDR): InterSense emulates the specified andwidth and latency values with less than 2% error between the expected and measured values. To demonstrate the InterSense's ease of use, we present a case study, where we apply InterSense for sensitivity analysis of four applications and benchmarks for getting non-trivial insights.

A common approach for improving application performance is to process its working set from memory. For datasets that do not fit into DRAM of a single machine this leads to a design of scale-out applications, where the application dataset is partitioned and processed by a cluster of machines. Performance of distributed memory applications, implemented using MPI (Message Passing Interface), inherently depends on performance of communication layer, which is largely defined by performance characteristics of underlying interconnect. During last couple years, many Big Data applications, e.g., Hadoop, Spark, Memcached, were re-written to take advantage of Remote Direct Memory Access (RDMA) technology and RDMA-capable interconnects which provide fast and high-bandwidth communications. The application analysis of potential performance improvements due to faster and higher bandwidth interconnects is a challenging task. Does the existing application implementation take a full advantage of the underlying interconnect or not? Will the application performance get worse if the interconnect has X% increased latency or Y% lower bandwidth? In this work, we introduce a novel emulation framework, called InterSense, which is implemented on top of existing high-speed interconnect, such as InfiniBand, and which provides two performance knobs for changing the interconnect bandwidth and latency. This approach offers an easy-to-use framework for a sensitivity analysis of complex distributed applications to communication layer performance instead of creating customized and time-consuming application models to answer the same questions. We evaluate the emulator accuracy with popular OSU MPI benchmarks: InterSense emulates the specified bandwidth and latency values with less than 2% error between the expected and measured values. We apply InterSense for sensitivity analysis of two new benchmarks, such as GUPS and Graph 500 to demonstrate the emulator's ease of use in getting non-trivial insights.

Next-generation non-volatile memory (NVM) technologies, such as phase-change memory and memristors, can enable computer systems infrastructure to continue keeping up with the voracious appetite of data-centric applications for large, cheap, and fast storage. Persistent memory has emerged as a promising approach to accessing emerging byte-addressable non-volatile memory through processor load/store instructions. Due to lack of commercially available NVM, system software researchers have mainly relied on emulation to model persistent memory performance. However, existing emulation approaches are either too simplistic, or too slow to emulate large-scale workloads, or require special hardware. To fill this gap and encourage wider adoption of persistent memory, we developed a performance emulator for persistent memory, called Quartz. Quartz enables an efficient emulation of a wide range of NVM latencies and bandwidth characteristics for performance evaluation of emerging byte-addressable NVMs and their impact on applications performance (without modifying or instrumenting their source code) by leveraging features available in commodity hardware. Our emulator is implemented on three latest Intel Xeon-based processor architectures: Sandy Bridge, Ivy Bridge, and Haswell. To assist researchers and engineers in evaluating design decisions with emerging NVMs, we extend Quartz for emulating the application execution on future systems with two types of memory: fast, regular volatile DRAM and slower persistent memory. We evaluate the effectiveness of our approach by using a set of specially designed memory-intensive benchmarks and real applications. The accuracy of the proposed approach is validated by running these programs both on our emulation platform and a multisocket (NUMA) machine that can support a range of memory latencies. We show that Quartz can emulate a range of performance characteristics with low overhead and good accuracy (with emulation errors 0.2% - 9%).

Exponential increase of online data and a corresponding growth of data-centric applications (Big Data analytics) forces system architects to revisit assumptions and requirements of the future system design. New non-volatile memory (NVM) technologies, such as Phase-Change Memory (PCM) and HP Memristor offer significantly improved latency and power efficiency compared to flash and hard drives. Many future systems are expected to have both DRAM and NVM. This can radically change system and software design, and enable new style of Big Data processing applications. However, the commercial unavailability of new NVMs technologies and uncertainty of their performance characteristics make it difficult to assess new system software stacks and to study their performance impact on future workloads. To bridge this gap and encourage an early design phase, we are building a DRAM-based performance emulation platform, called NVMpro, that leverages features available in commodity hardware, to emulate different latency and bandwidth characteristics of future NVM technologies. NVMpro enables an efficient and accurate emulation of a wide range of NVM latencies and bandwidth characteristics for performance evaluation of emerging byte-addressable NVMs and their impact on applications performance without modifying or instrumenting their source code.

Hadoop and its variants have been widely used for processing large scale analytics tasks in a cluster environment. However, use of a commodity cluster for analytics tasks needs to be reconsidered based on two key observations: (1) in recent years, large memory, multicore machines have become more affordable; and (2) recent studies show that most analytics tasks in practice are smaller than 100 GB. Thus, replacing a commodity cluster with a large memory, multicore machine can enable in-memory analytics at an affordable cost. However= programming on a big-memory, multicore machine is a challenge. Multi-threaded programming is notoriously difficult. Further, the memory design of most large memory servers follows non-uniform memory access (NUMA) architecture. While NUMA-aware programming often leads to high efficiency in analytics tasks, it is usually done in an ad hoc manner.

Regulatory policies such as EU privacy, HIPAA, and PCI-DSS place requirements on availability, integrity, migration, retention, and access of data, and compliance with such policies on stored data remains a key hurdle to cloud computing. This paper proposes a policy management service that offers scalable management of data retention policies attached to data objects stored in a cloud environment. An important aspect of any data retention service is permanent deletion of data. We achieve secure data deletion by encrypting the data when stored, and then deleting the encryption key at a specified retention time. Thus, we effectively delete the data object and its copies stored in online and offline environments. Our data retention service includes a highly scalable and secure encryption key store to manage encryption keys on-line. A prototype deployed on a 16-machine Linux cluster currently supports 56 MB/sec for encryption, 76 MB/sec for decryption, 31,000 retention policies/sec read and 15,000 retention policies/sec write.

The design, deployment and execution of business process models and their associated security models is expensive and time consuming. This is because these activities usually involve multiple stakeholders that include business domain experts, security experts, web service developers and IT operations teams, and there is no streamlined development environment to allow these stakeholders to work collaboratively on a business process. We have developed a cloud-based model-driven development and execution environment called SSC4Cloud to provide a shared business process modeling workspace and a business process execution environment. More specifically, with the shared modeling workspace, business process models can be developed, refined and shared. Within the shared execution environment, a business process model is translated into a WS-BPEL based executable model, which is then assigned for execution in a virtual machine container from a shared machine cluster. The common model execution environment supports both business process execution and enforcement of the security requirements attached to the business process models.

Compliance with regulatory policies on data remains a key hurdle to cloud computing. Policies such as EU privacy, HIPAA, and PCI-DSS place requirements on data availability, integrity, migration, retention, and access, among many others. This paper proposes a policy management service that offers scalable management of data retention policies attached to data objects stored in a cloud environment. The management service includes a highly available and secure encryption key store to manage the encryption keys of data objects. By deleting the encryption key at a specified retention time associated with the data object, we effectively delete the data object and its copies stored in online and offline environments. To achieve scalability, our service uses Hadoop MapReduce to perform parallel management tasks, such as data encryption and decryption, key distribution and retention policy enforcement. A prototype deployed in a 16-machine Linux cluster currently supports 56 MB/sec for encryption, 76 MB/sec for decryption, 31,000 retention policies/sec read and 15,000 retention policies/sec write.

Cloud Computing, Business Process Modeling (BPM) and Service-oriented architectures (SOA) are playing a relevant role in the evolution of Information Technology (IT). A considerable number of system developers are using Cloud technologies to deploy and make available systems over the Internet. Business Process Management standards are being widely used to model business requirements. In addition, SOA-based systems are considered an interesting approach to execute high-level business process specifications. Based on the fact that business processes are executed, usually, using services available in network environments, security requirements should be considered, especially when dealing with sensitive data (e.g., credit card information or personal data). Despite the increasing need for specifying security mechanisms in web service compositions in the Cloud, this topic remains a challenge for many reasons, including the known difficulty of expressing security requirements at a business level and the enforcement of such requirements at an execution level in a cloud environment. This work presents an environment to collaboratively model business processes considering security requirements and to automatically deploy them in the Cloud with security requirements enforcement. The business process is realized through the utilization of web service composition. This environment consists of a set of tools to support the business process modeling and secure service composition execution in the Cloud. Security-related information can be shared among different users in the Cloud and used to enable the activation and configuration of security mechanisms. The proposed approach is showcased in a Virtual Travel Agency scenario to show its feasibility.

Many cloud service providers offer outsourcing capabilities to businesses using the software-as-a-service delivery model. In this delivery model, sensitive business data need to be stored and processed outside the control of the business. The ability to manage data in compliance with regulatory and corporate policies, which we refer to as data assurance, is an essential success factor for this delivery model. There exist challenges to express service data assurance capabilities, capture customers' requirements, and enforce these policies inside service providers' environments. This paper addresses these challenges by proposing Global Enforcement Of Data Assurance Controls (GEODAC), a policy framework that enables the expression of both service providers' capabilities and customers' requirements, and enforcement of the agreed-upon data assurance policies in service providers' environments. High-level policy statements are backed in the service environment with a state machine-based representation of policies in which each state represents a data lifecycle stage. Data assurance policies that define requirements on data retention, data migration, data appropriateness for use, etc. can be described and enforced. The approach has been implemented in a prototype tool and evaluated in a services environment.

A key goal of the Services Oriented Architecture is the composition of independently written and managed services. However, managing access to these services has proven to be a problem. A particularly difficult case involves a service that invokes another service to satisfy an initial request. In a number of cases, implementations are able to achieve either the desired functionality or the required security, but not both at the same time. We say that this service composition suffers from the transitive access problem. We show that the problem arises from a poor choice of access control mechanism, one that uses subject authentication to make access decisions, and that the problem does not occur if we use delegatable authorizations.

The loss of customer account information makes headlines all too often. The Payment Card Industry (PCI) Data Security Standard (DSS) has generated resistance from merchants because of its cost and complexity. Unfortunately, innovating in the Point of Sale (PoS) space is hard because the protocols have been standardized, tightly constraining new ideas. Using mobile phones as payment devices opens up some opportunities, but any solution must minimize changes in the back-end processing. This paper describes a Near Field Communication (NFC) based mobile phone payment solution in a service-oriented environment, which provides the needed data protection without requiring costly changes to the payment processing infrastructure. In the payment transaction path, a personal mobile phone is viewed as a single, user-trusted touch point. Compared to other solutions available, our approach better protects user credentials, provides better user control over the transaction, and supports both proximity and remote transactions. The prototype has been integrated with an instore kiosk application and a HP multi-channel banking platform to demonstrate its value in the retail environment.

Large numbers of consumers, businesses, and public entities are now using the Internet for a variety of transactions. This has enabled service providers to offer outsourcing capabilities to business customers using software-as-a-service delivery models in services marketplaces. However, challenges remain in widespread acceptance of such delivery models because they require customers to share business critical data with the service providers. This paper presents a policy framework that enables businesses to communicate data management policies with service providers at an arbitrarily granular level. Policy is described as a state machine with each state representing a lifecycle stage, and attached to data when it is shared between services. Data management related policies including data appropriateness, data quality assurance, data retention and data migration can be described in this framework and enforced correspondingly.

The mobile shopping assistant (MSA) is a mobile application platform to deliver real-time, in-store, and personalized services, such as personalized product offerings and in-store customer advisory support, to improve the shopping experiences of in-store customers. The service delivery network that powers the MSA involves retail stores and their business partners such as manufacturers. This paper presents the core technologies that we developed in this cross-organizational service network to support the MSA and its personalized services, with focus on service delivery, customer behavior understanding and information sharing. Our event-based techniques allow customers, stores and manufacturers to deliver and consume the services in a loosely coupled manner, thus solving a critical store-specific real-time engagement problem in a seamless way. Service response tracking enables the stores to construct a comprehensive view of a customerpsilas in-store shopping behavior. Finally, the cross-organizational authorization-based access control mechanism effectively enforces information sharing between the stores and their partners.

It is extremely hard for a global organization with services over multiple channels to capture a consistent and unified view of its data, services, and interactions. While SOA and web services are addressing integration and interoperability problems, it is painful for an operational organization with legacy systems to quickly switch to service-based methods. We need methods to combine advantages of traditional (i.e. web, desktop, or mobile) application development environments and service-based deployments.In this paper, we focus on the design and implementation of session management as a core service to support business processes and go beyond application-specific sessions and web sessions. We develop local session components for different platforms and complement them with a remote "session service" that is independent of applications and platforms. We aim to close the gap between the two worlds by combining their performance, availability and interoperability advantages.

A new approach to testing component-based applications is presented, which exploits the practice in component-based systems of generating stub/skeleton modules and using these stubs/skeletons to construct a global perspective of end-to-end causality of inter-component communication. This global causality is captured regardless of reentrancy, callbacks, thread and process boundaries, and unsynchronized clocks. The captured logs created from the interception points are used to construct a system-wide component interaction model that can expose the inter-component dependencies usually hidden in static analysis of application code. These discovered dependencies are used to create a test boundary for applying a component test harness for that component and the set of dependent components. Similarly, the discovered dependencies can be applied to pruning the available test cases to identify those cases that are best suited to exposing defects when one or more components are changed. A particular advantage of the approach has been the ability to isolate the sequence of events that led up to a crash or a deadlock condition and view the entire system behavior (not just a particular thread's perspective or a linear log of intercepted messages).

Current software development techniques and tools lack the capability to characterize function call chains in multithreaded and distributed applications built upon component technologies like CORBA, COM and J2EE. The root cause is that causal linkage information necessary to trace end-to-end call chains is private to each vendor's runtime and often unavailable for logging or analysis. We propose and demonstrate a mechanism for maintaining and correlating global causality information of component-based applications, and using this information to expose and characterize function call chains and their associated behaviors in such multithreaded and distributed applications. Our approach relies on a global virtual tunnel facilitated by the instrumented stubs and skeletons. This tunnel maintains and correlates causal information throughout the end-to-end call chains spanning threads, processes and processors. As a result, monitoring data captured locally can be correlated and system-wide propagation of timing latency and CPU utilization becomes perceivable.

Real time control systems typically receive streams of sampled data signals, process them, and generate output to control devices via actuators. Control system components make assumptions about sampling rate and other characteristics, such as filtering of data streams. These application-specific semantic dependencies and time-sensitive semantic properties are usually not explicit in the source code implementation. The paper discusses an approach to modeling the system architecture of real time applications, capturing semantic dependencies, and recording time-sensitive properties. Analysis of the model can identify inconsistencies in a system design and can determine the impact of a system change. This architectural dependency model can evolve incrementally to record hidden side effects, as they are discovered during maintenance, and prevent such errors in the future.