Richard A. Lethin

Efficient and scalable computations with sparse tensors

For applications that deal with large amounts of high dimensional multi-aspect data, it becomes natural to represent such data as tensors or multi-way arrays. Multi-linear algebraic computations such as tensor decompositions are performed for summarization and analysis of such data. Their use in real-world applications can span across domains such as signal processing, data mining, computer vision, and graph analysis. The major challenges with applying tensor decompositions in real-world applications are (1) dealing with large-scale high dimensional data and (2) dealing with sparse data. In this paper, we address these challenges in applying tensor decompositions in real data analytic applications. We describe new sparse tensor storage formats that provide storage benefits and are flexible and efficient for performing tensor computations. Further, we propose an optimization that improves data reuse and reduces redundant or unnecessary computations in tensor decomposition algorithms. Furthermore, we couple our data reuse optimization and the benefits of our sparse tensor storage formats to provide a memory-efficient scalable solution for handling large-scale sparse tensor computations. We demonstrate improved performance and address memory scalability using our techniques on both synthetic small data sets and large-scale sparse real data sets.

High-performance many-core networking: design and implementation

Operating systems play a key role in providing general purpose services to upper layer applications at the highest available performance level. The two design requirements --- generality and performance --- are however in contention: the more general purpose a service layer is, the more overhead it incurs in accessing domain-specific high-performance features provided by the layers beneath it. This trade-off comes to manifest in modern computer systems as the state-of-the-art has evolved from architectures with a few number of cores to systems employing a very large number of cores (many-core systems). Such evolution has rendered the networking layer in current operating systems inefficient as its general purpose design deprives it from a proper use of the large number of cores. In this paper we introduce DNAC (Dynamic Network Acceleration for Many-Core), a high-performance abstraction layer designed to target the maximum network performance available from the network interface in many-core general purpose operating systems.

Cyber Security through Multidimensional Data Decompositions

Traditional machine learning approaches are plagued with problems for practical use in operational cyber security. The class of unsupervised learning algorithms called tensor decompositions provide a new approach for analyzing network traffic data that avoids these traditional problems. Tensors are a natural representation for multidimensional data as an array with arbitrary dimensions. Tensor decompositions factor the data into components, each of which represents a different pattern of activity from within the original data.We use ENSIGN, a tensor decomposition toolbox developed by Reservoir Labs, in the security operations center for the SCinet network at SC15 - The International Conference for High Performance Computing, Networking, Storage and Analysis. ENSIGN integrates naturally into an operational cyber security framework by extracting anomalous patterns of network traffic. In this paper, we present two case studies highlighting specific actionable results: one, discovering an external attacker and tracing the evolution of the attack over time, and the other, extracting an example of data exfiltration that the actor disguised as DNS activity and cleanly separating it from normal DNS activity. Through proof-of-concept experiments at SC15, we successfully demonstrate concrete and practical use of ENSIGN and make a critical step forward towards delivering an integrated tensor analysis engine for network security.

Topic modeling for analysis of big data tensor decompositions

Tensor decompositions are a class of algorithms used for unsupervised pattern discovery. Structured, multidimensional datasets are encoded as tensors and decomposed into discrete, coherent patterns captured as weighted collections of high-dimensional vectors known as components. Tensor decompositions have recently shown promising results when addressing problems related to data comprehension and anomaly discovery in cybersecurity and intelligence analysis. However, analysis of Big Data tensor decompositions is currently a critical bottleneck owing to the volume and variety of unlabeled patterns that are produced. We present an approach to automated component clustering and classification based on the Latent Dirichlet Allocation (LDA) topic modeling technique and show example applications to representative cybersecurity and geospatial datasets.

Automatic code generation and data management for an asynchronous task-based runtime

Hardware scaling and low-power considerations associated with the quest for exascale and extreme scale computing are driving system designers to consider new runtime and execution models such as the event-driven-task (EDT) models that enable more concurrency and reduce the amount of synchronization. Further, for performance, productivity, and code sustainability reasons, there is an increasing demand for auto-parallelizing compiler technologies to automatically produce code for EDT-based runtimes. However achieving scalable performance in extreme-scale systems with auto-generated codes is a non-trivial challenge. Some of the key requirements that are important for achieving good scalable performance across many EDT-based systems are: (1) scalable dynamic creation of task-dependence graph and spawning of tasks, (2) scalable creation and management of data and communications, and (3) dynamic scheduling of tasks and movement of data for scalable asynchronous execution. In this paper, we develop capabilities within R-Stream - an automatic source-to-source optimization compiler - for automatic generation and optimization of code and data management targeted towards Open Community Runtime (OCR) - an exascale-ready asynchronous task-based runtime. We demonstrate the effectiveness of our techniques through performance improvements on various benchmarks and proxy application kernels that are relevant to the extreme-scale computing community.