Big data has been revolutionizing individualized medicine, improving the results of diagnostic imaging, genetic testing, and by providing frameworks for electronic health record sharing and analysis. In this paper we present a nextstep in personalized data-driven health by demonstrating the capability of predictive individualized models to model peak postprandial plasma glucose concentrations. Past research has shown that postprandial plasma glucose concentrations are highly predictive of both the development of type 2 diabetes, β-cell damage in the pancreas due to glucose toxicity, and a wide range of cardiovascular complications. We present a framework for individualized data collection, model parameterization, and prediction which allows for ongoing estimation of postprandial glucose after short training windows to aid in determining when an individual may be at risk for developing pre-diabetes, type 2 diabetes, or complications due to high peak postprandial blood glucose.

Summary Lack of trust has become one of the main concerns of users who tend to utilize one or multiple Cloud providers. Trustworthy Cloud‐based computing and data storage require secure and efficient solutions which allow clients to remotely store and process their data in the Cloud. User‐side encryption is an established method to secure the user data on the Cloud. However, using encryption, we lose processing capabilities, such as searching, over the Cloud data. In this paper, we present RESeED, a tool that provides user‐transparent and Cloud‐agnostic regular‐expression search functionality over encrypted data across multiple Clouds. Upon a client's intent to upload a new document to the Cloud, RESeED analyzes the document's content and updates its data structures accordingly. Then, it encrypts and transfers the document to the Cloud. RESeED provides the regular‐expression search functionality over encrypted data by translating the search queries on‐the‐fly to finite automata and analyzing concise and secure representations of the data before asking the Cloud to download the encrypted documents. RESeED's parallel architecture enables efficient search over large‐scale (and potentially big data scale) data‐sets. We evaluate the performance of RESeED experimentally and demonstrate its scalability and correctness using real‐world data‐sets fromarXiv.organd Internet Engineering Task Force (IETF). Our results show that RESeED produces accurate query responses with a reasonable (≃6%) storage overhead. The results also demonstrate that for many search queries, RESeED performs faster in compare with thegreputility that functions on unencrypted data. Copyright © 2017 John Wiley & Sons, Ltd.

We propose new algorithms for implementing a software-defined data center (SDDC) to improve the dependability of storage systems without the addition of new hardware. We define the construction of a system that can predict its future resource requirements and act on these predictions to allocate overprovisioned resources to improve reliability. We introduce algorithms for implementing a smart SDDC (SSDDC) that characterizes user I/O transactions (writes and deletes), and use these models to predict the level of overprovisioning within a system, overbooking excess resources to improve reliability, while mitigating the impact on quality of service. We compare several implementations of our methods experimentally, and discuss methods for improving the fault-tolerance of our S2DDC, present experimental results showcasing our ability to improve system reliability showing the decrease in expected annual block loss due to disk failures and latent sector errors, and highlight the benefit of dependence based usage models in estimating overprovisioning.

Lead poisoning is a major public health problem that affects hundreds of thousands of children in the United States every year. A common approach to identifying lead hazards is to test all children for elevated blood lead levels and then investigate and remediate the homes of children with elevated tests. This can prevent exposure to lead of future residents, but only after a child has been poisoned. This paper describes joint work with the Chicago Department of Public Health (CDPH) in which we build a model that predicts the risk of a child to being poisoned so that an intervention can take place before that happens. Using two decades of blood lead level tests, home lead inspections, property value assessments, and census data, our model allows inspectors to prioritize houses on an intractably long list of potential hazards and identify children who are at the highest risk. This work has been described by CDPH as pioneering in the use of machine learning and predictive analytics in public health and has the potential to have a significant impact on both health and economic outcomes for communities across the US.

Capabilities for trustworthy cloud-based computing and data storage require usable, secure and efficient solutions which allow clients to remotely store and process their data in the cloud. In this paper, we present RESeED, a tool which provides user-transparent and cloud-agnostic search over encrypted data using regular expressions without requiring cloud providers to make changes to their existing infrastructure. When a client asks RESeED to upload a new file in the cloud, RESeED analyzes the file's content and updates novel data structures accordingly, encrypting and transferring the new data to the cloud. RESeED provides regular expression search over this encrypted data by translating queries on-the-fly to finite automata and analyzes efficient and secure representations of the data before asking the cloud to download the encrypted files. We evaulate a working prototype of RESeED experimentally (currently publicly available) and show the scalability and correctness of our approach using real-world data sets from arXiv.org and the IETF. We show absolute accuracy for RESeED, with very low (6%) overhead, and high performability, even beating grep for some benchmarks.

We present Reseed, a tool that provides user-transparent and Cloud-agnostic regular expression search over encrypted data without requiring trust in the Cloud, or changes to Cloud infrastructure. Upon receiving a search query, Reseed translates it to a finite automata and analyzes efficient and secure representations of the data before asking the Cloud to download the matching encrypted files. We demonstrate and evaluate a working prototype of Reseed and show the scalability and correctness of our approach using data from arXiv.org. For more details see our Technical Report.

Trustworthy cloud services require practical secure storage and data processing techniques that enable end-users to upload sensitive data and perform computations remotely without having to first download the data. In this paper, we present Dragon fruit, a cloud provider-agnostic searchable cloud data storage solution, that allows the utilization of several existing cloud providers, and the execution of search queries over encrypted data by customers, using unmodified cloud infrastructures. In particular, Dragon fruit makes use of filename search capabilities that most existing cloud providers support to embed searchable data within the cloud. Dragon fruit supports queries using complex Boolean expressions on structured data formats such as JSON. Our evaluation on a real-world test-bed show that Dragon fruit is able to work with several cloud providers such as Google-Drive and Drop box simultaneously and is able to respond to search requests within a few seconds proving a reasonable performance overhead for practical usage.

As both the amount of data to be stored and the rate of data production grows, data center designers and operators face the challenge of planning and managing systems whose characteristics, workloads, and performance, availability, and reliability goals change rapidly. As we move towards software-defined data centers (SDDCs) the ability to reconfigure and adapt our solutions is increasing, but to take full advantage of that increase we must design smarter, more intelligent systems that are aware of how they are being used and able to deliver accurate predictions of their characteristics, workloads, and goals. In this paper we propose a novel algorithm for use in an intelligent, user-aware SDDC which performs run-time analysis of user storage system activity in a manner that has a minimal impact on performance and provides accurate estimations of future user activity.

In this paper we present a framework for analyzing the fault tolerance of deduplicated storage systems. We discuss methods for building models of deduplicated storage systems by analyzing empirical data on a file category basis. We provide an algorithm for generating component-based models from this information and a specification of the storage system architecture. Given the complex nature of detailed models of deduplicated storage systems, finding a solution using traditional discrete event simulation or numerical solvers can be difficult. We introduce an algorithm which allows for a more efficient solution by exploiting the underlying structure of dependencies to decompose the model of the storage system. We present a case study of our framework for a real system.We analyze a production deduplicated storage system and propose extensions which improve fault tolerance while maintaining high storage efficiency.

Modern storage systems are employing data deduplication with increasing frequency. Often the storage systems on which these techniques are deployed contain important data, and utilize fault-tolerant hardware and software to improve the reliability of the system and reduce data loss. We suggest that data deduplication introduces inter-file relationships that may have a negative impact on the fault tolerance of such systems by creating dependencies that can increase the severity of data loss events. We present a framework composed of data analysis methods and a model of data deduplication that is useful in studying the reliability impact of data deduplication. The framework is useful for determining a deduplication strategy that is estimated to satisfy a set of reliability constraints supplied by a user.

Mobius 2.3 is an extensible dependability, security, and performance modeling environment for large-scale discrete-event systems. It provides multiple model formalisms and solution techniques, facilitating the representation of each part of a system in the formalism that is most appropriate for it, and the application of the solution method or methods best-suited to estimating the system's behavior. Since its initial release in 2001, many advances have been made in Moumlbius's design and implementation that have strengthened its place in the modeling and analysis community. With almost a decade of widespread academic and industrial use, Moumlbius has proven itself to be useful in a wide variety of modeling situations. This paper documents the current feature set of Mobius 2.3, emphasizing recent significant enhancements.

Despite the reliability of modern disks, recent studies have made it clear that a new class of faults, UndetectedDisk Errors (UDEs) also known as silent data corruption events, become a real challenge as storage capacity scales. While RAID systems have proven effective in protecting data from traditional disk failures, silent data corruption events remain a significant problem unaddressed by RAID. We present a fault model for UDEs, and a hybrid framework for simulating UDEs in large-scale systems. The framework combines a multi-resolution discrete event simulator with numerical solvers. Our implementation enables us to model arbitrary storage systems and workloads and estimate the rate of undetected data corruptions. We present results for several systems and workloads, from gigascale to petascale. These results indicate that corruption from UDEs is a significant problem in the absence of protection schemes and that such schemes dramatically decrease the rate of undetected data corruption.

Petascale computing requires I/O subsystems that can keep up with the dramatic computing power demanded by such systems. TOP500.org ranks top computers based on their peak compute performance, but there has not been adequate investigation of the current state-of-the-art and future requirements of storage area networks that support petascale computers. Dependable scaling of an I/O subsystem to support petascale computing is not as simple as adding more storage servers. In this paper, we present a stochastic activity network model that uses failure rates computed from real logs to predict the reliability and availability of the storage architecture of the Abe cluster at the National Center for Supercomputing Applications (NCSA). We then use the model to evaluate the challenges encountered as one scales the number of storage servers to support petascale computing. The results present new insights regarding the dependability challenges that will be encountered when building next-generation petabyte storage. Furthermore, we provide insight into a new design approach that will enable system designers to integrate the trace-based analysis of parameter values from real system data into their stochastic models to allow informed design choices.