Search

There is a tremendous amount of interest in AI/ML technologies due to the proliferation of generative AI applications such as ChatGPT. This trend has significantly increased demand on GPUs, which are the workhorses for training AI models. Due to the high costs of GPUs and lacking supply, it has become of interest to optimize GPU usage in HPC centers. MIT Lincoln Laboratory Supercomputing Center (LLSC) has developed an easy-to-use GPU sharing feature supported by LLSC-developed tools including LLsub and LLMapReduce. This approach overcomes some of the limitations with the existing methods for GPU sharing. This allows users to apply GPU sharing whenever possible while they are developing their AI/ML models and/or doing parametric study on their AI models or executing other GPU applications. Based on our initial experimental results with GPU sharing, GPU sharing with triples mode is easy to use and achieved significant improvement in GPU usage and throughput performance for certain types of AI applications.

The increasing use and cost of high performance computing (HPC) requires new easy-to-use tools to enable HPC users and HPC systems engineers to transparently understand the utilization of resources. The MIT Lincoln Laboratory Supercomputing Center (LLSC) has developed a simple command, LLload, to monitor and characterize HPC workloads. LLload plays an important role in identifying opportunities for better utilization of compute resources. LLload can be used to monitor jobs both programmatically and interactively. LLload can characterize users' jobs using various LLload options to achieve better efficiency. This information can be used to inform the user to optimize HPC workloads and improve both CPU and GPU utilization. This includes improvements using judicious oversubscription of the computing resources. Preliminary results suggest significant improvement in GPU utilization and overall throughput performance with GPU overloading in some cases. By enabling users to observe and fix incorrect job submission and/or inappropriate execution setups, LLload can increase the resource usage and improve the overall throughput performance. LLload is a light-weight, easy-to-use tool for both HPC users and HPC systems engineers to monitor HPC workloads to improve system utilization and efficiency.

HPC systems used for research run a wide variety of software and workflows. This software is often written or modified by users to meet the needs of their research projects, and rarely is built with security in mind. In this paper we explore several of the key techniques that MIT Lincoln Laboratory Supercomputing Center has deployed on its systems to manage the security implications of these workflows by providing enforced separation for processes, filesystem access, network traffic, and accelerators to make every user feel like they are running on a personal HPC.

The MIT/IEEE/Amazon GraphChallenge encourages community approaches to developing new solutions for analyzing graphs and sparse data derived from social media, sensor feeds, and scientific data to discover relationships between events as they unfold in the field. The anonymized network sensing Graph Challenge seeks to enable large, open, community-based approaches to protecting networks. Many large-scale networking problems can only be solved with community access to very broad data sets with the highest regard for privacy and strong community buy-in. Such approaches often require community-based data sharing. In the broader networking community (commercial, federal, and academia) anonymized source-to-destination traffic matrices with standard data sharing agreements have emerged as a data product that can meet many of these requirements. This challenge provides an opportunity to highlight novel approaches for optimizing the construction and analysis of anonymized traffic matrices using over 100 billion network packets derived from the largest Internet telescope in the world (CAIDA). This challenge specifies the anonymization, construction, and analysis of these traffic matrices. A GraphBLAS reference implementation is provided, but the use of GraphBLAS is not required in this Graph Challenge. As with prior Graph Challenges the goal is to provide a well-defined context for demonstrating innovation. Graph Challenge participants are free to select (with accompanying explanation) the Graph Challenge elements that are appropriate for highlighting their innovations., Comment: Accepted to IEEE HPEC 2024

Understanding what is normal is a key aspect of protecting a domain. Other domains invest heavily in observational science to develop models of normal behavior to better detect anomalies. Recent advances in high performance graph libraries, such as the GraphBLAS, coupled with supercomputers enables processing of the trillions of observations required. We leverage this approach to synthesize low-parameter observational models of anonymized Internet traffic with a high regard for privacy., Comment: Accepted to IEEE HPEC, 7 pages, 6 figures, 1 table, 41 references

One of the more complex tasks for researchers using HPC systems is performance monitoring and tuning of their applications. Developing a practice of continuous performance improvement, both for speed-up and efficient use of resources is essential to the long term success of both the HPC practitioner and the research project. Profiling tools provide a nice view of the performance of an application but often have a steep learning curve and rarely provide an easy to interpret view of resource utilization. Lower level tools such as top and htop provide a view of resource utilization for those familiar and comfortable with Linux but a barrier for newer HPC practitioners. To expand the existing profiling and job monitoring options, the MIT Lincoln Laboratory Supercomputing Center created LLoad, a tool that captures a snapshot of the resources being used by a job on a per user basis. LLload is a tool built from standard HPC tools that provides an easy way for a researcher to track resource usage of active jobs. We explain how the tool was designed and implemented and provide insight into how it is used to aid new researchers in developing their performance monitoring skills as well as guide researchers in their resource requests.

Recent analyses combining advanced theoretical techniques and high-quality data from thousands of simultaneously recorded neurons provide strong support for the hypothesis that neural dynamics operate near the edge of instability across regions in the brain. However, these analyses, as well as related studies, often fail to capture the intricate temporal structure of brain activity as they primarily rely on time-integrated measurements across neurons. In this study, we present a novel framework designed to explore signatures of criticality across diverse frequency bands and construct a much more comprehensive description of brain activity. Additionally, we introduce a method for projecting brain activity onto a basis of spatio-temporal patterns, facilitating time-dependent dimensionality reduction. Applying this framework to a magnetoencephalography dataset, we observe significant differences in both criticality signatures and spatio-temporal activity patterns between healthy subjects and individuals with Parkinson's disease.

Advancing our knowledge of how the brain processes information remains a key challenge in neuroscience. This thesis combines three different approaches to the study of the dynamics of neural networks and their encoding representations: a computational approach, that builds upon basic biological features of neurons and their networks to construct effective models that can simulate their structure and dynamics; a machine-learning approach, which draws a parallel with the functional capabilities of brain networks, allowing us to infer the dynamical and encoding properties required to solve certain input-processing tasks; and a final, theoretical treatment, which will take us into the fascinating hypothesis of the "critical" brain as the mathematical foundation that can explain the emergent collective properties arising from the interactions of millions of neurons. Hand in hand with physics, we venture into the realm of neuroscience to explain the existence of quasi-universal scaling properties across brain regions, setting out to quantify the distance of their dynamics from a critical point. Next, we move into the grounds of artificial intelligence, where the very same theory of critical phenomena will prove very useful for explaining the effects of biologically-inspired plasticity rules in the forecasting ability of Reservoir Computers. Halfway into our journey, we explore the concept of neural representations of external stimuli, unveiling a surprising link between the dynamical regime of neural networks and the optimal topological properties of such representation manifolds. The thesis ends with the singular problem of representational drift in the process of odor encoding carried out by the olfactory cortex, uncovering the potential synaptic plasticity mechanisms that could explain this recently observed phenomenon., Comment: A dissertation submitted to the University of Granada in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Expanding the scientific tools available to protect computer networks can be aided by a deeper understanding of the underlying statistical distributions of network traffic and their potential geometric interpretations. Analyses of large scale network observations provide a unique window into studying those underlying statistics. Newly developed GraphBLAS hypersparse matrices and D4M associative array technologies enable the efficient anonymized analysis of network traffic on the scale of trillions of events. This work analyzes over 100,000,000,000 anonymized packets from the largest Internet telescope (CAIDA) and over 10,000,000 anonymized sources from the largest commercial honeyfarm (GreyNoise). Neither CAIDA nor GreyNoise actively emit Internet traffic and provide distinct observations of unsolicited Internet traffic (primarily botnets and scanners). Analysis of these observations confirms the previously observed Cauchy-like distributions describing temporal correlations between Internet sources. The Gull lighthouse problem is a well-known geometric characterization of the standard Cauchy distribution and motivates a potential geometric interpretation for Internet observations. This work generalizes the Gull lighthouse problem to accommodate larger classes of coastlines, deriving a closed-form solution for the resulting probability distributions, stating and examining the inverse problem of identifying an appropriate coastline given a continuous probability distribution, identifying a geometric heuristic for solving this problem computationally, and applying that heuristic to examine the temporal geometry of different subsets of network observations. Application of this method to the CAIDA and GreyNoise data reveals a several orders of magnitude difference between known benign and other traffic which can lead to potentially novel ways to protect networks., Comment: 9 pages, 7 figures, IEEE HPEC 2023 (accepted)

pPython seeks to provide a parallel capability that provides good speed-up without sacrificing the ease of programming in Python by implementing partitioned global array semantics (PGAS) on top of a simple file-based messaging library (PythonMPI) in pure Python. pPython follows a SPMD (single program multiple data) model of computation. pPython runs on a single-node (e.g., a laptop) running Windows, Linux, or MacOS operating systems or on any combination of heterogeneous systems that support Python, including on a cluster through a Slurm scheduler interface so that pPython can be executed in a massively parallel computing environment. It is interesting to see what performance pPython can achieve compared to the traditional socket-based MPI communication because of its unique file-based messaging implementation. In this paper, we present the point-to-point and collective communication performances of pPython and compare them with those obtained by using mpi4py with OpenMPI. For large messages, pPython demonstrates comparable performance as compared to mpi4py., Comment: arXiv admin note: substantial text overlap with arXiv:2208.14908

Matrix/array analysis of networks can provide significant insight into their behavior and aid in their operation and protection. Prior work has demonstrated the analytic, performance, and compression capabilities of GraphBLAS (graphblas.org) hypersparse matrices and D4M (d4m.mit.edu) associative arrays (a mathematical superset of matrices). Obtaining the benefits of these capabilities requires integrating them into operational systems, which comes with its own unique challenges. This paper describes two examples of real-time operational implementations. First, is an operational GraphBLAS implementation that constructs anonymized hypersparse matrices on a high-bandwidth network tap. Second, is an operational D4M implementation that analyzes daily cloud gateway logs. The architectures of these implementations are presented. Detailed measurements of the resources and the performance are collected and analyzed. The implementations are capable of meeting their operational requirements using modest computational resources (a couple of processing cores). GraphBLAS is well-suited for low-level analysis of high-bandwidth connections with relatively structured network data. D4M is well-suited for higher-level analysis of more unstructured data. This work demonstrates that these technologies can be implemented in operational settings., Comment: Accepted to IEEE HPEC, 8 pages, 8 figures, 1 table, 69 references. arXiv admin note: text overlap with arXiv:2203.13934. text overlap with arXiv:2309.01806

Defending community-owned cyber space requires community-based efforts. Large-scale network observations that uphold the highest regard for privacy are key to protecting our shared cyberspace. Deployment of the necessary network sensors requires careful sensor placement, focusing, and calibration with significant volumes of network observations. This paper demonstrates novel focusing and calibration procedures on a multi-billion packet dataset using high-performance GraphBLAS anonymized hypersparse matrices. The run-time performance on a real-world data set confirms previously observed real-time processing rates for high-bandwidth links while achieving significant data compression. The output of the analysis demonstrates the effectiveness of these procedures at focusing the traffic matrix and revealing the underlying stable heavy-tail statistical distributions that are necessary for anomaly detection. A simple model of the corresponding probability of detection ($p_{\rm d}$) and probability of false alarm ($p_{\rm fa}$) for these distributions highlights the criticality of network sensor focusing and calibration. Once a sensor is properly focused and calibrated it is then in a position to carry out two of the central tenets of good cybersecurity: (1) continuous observation of the network and (2) minimizing unbrokered network connections., Comment: Accepted to IEEE HPEC, 9 pages, 12 figures, 1 table, 63 references, 2 appendices

Deciphering the underpinnings of the dynamical processes leading to information transmission, processing, and storing in the brain is a crucial challenge in neuroscience. An inspiring but speculative theoretical idea is that such dynamics should operate at the brink of a phase transition, i.e., at the edge between different collective phases, to entail a rich dynamical repertoire and optimize functional capabilities. In recent years, research guided by the advent of high-throughput data and new theoretical developments has contributed to making a quantitative validation of such a hypothesis. Here we review recent advances in this field, stressing our contributions. In particular, we use data from thousands of individually recorded neurons in the mouse brain and tools such as a phenomenological renormalization group analysis, theory of disordered systems, and random matrix theory. These combined approaches provide novel evidence of quasi-universal scaling and near-critical behavior emerging in different brain regions. Moreover, we design artificial neural networks under the reservoir-computing paradigm and show that their internal dynamical states become near critical when we tune the networks for optimal performance. These results not only open new perspectives for understanding the ultimate principles guiding brain function but also towards the development of brain-inspired, neuromorphic computation.

Python has become a standard scientific computing language with fast-growing support of machine learning and data analysis modules, as well as an increasing usage of big data. The Dynamic Distributed Dimensional Data Model (D4M) offers a highly composable, unified data model with strong performance built to handle big data fast and efficiently. In this work we present an implementation of D4M in Python. $D4M.py$ implements all foundational functionality of D4M and includes Accumulo and SQL database support via Graphulo. We describe the mathematical background and motivation, an explanation of the approaches made for its fundamental functions and building blocks, and performance results which compare $D4M.py$'s performance to D4M-MATLAB and D4M.jl., Comment: 8 pages, 7 figures, accepted to HPEC 2022

pPython seeks to provide a parallel capability that provides good speed-up without sacrificing the ease of programming in Python by implementing partitioned global array semantics (PGAS) on top of a simple file-based messaging library (PythonMPI) in pure Python. The core data structure in pPython is a distributed numerical array whose distribution onto multiple processors is specified with a map construct. Communication operations between distributed arrays are abstracted away from the user and pPython transparently supports redistribution between any block-cyclic-overlapped distributions in up to four dimensions. pPython follows a SPMD (single program multiple data) model of computation. pPython runs on any combination of heterogeneous systems that support Python, including Windows, Linux, and MacOS operating systems. In addition to running transparently on single-node (e.g., a laptop), pPython provides a scheduler interface, so that pPython can be executed in a massively parallel computing environment. The initial implementation uses the Slurm scheduler. Performance of pPython on the HPC Challenge benchmark suite demonstrates both ease of programming and scalability., Comment: arXiv admin note: substantial text overlap with arXiv:astro-ph/0606464

The brain is in a state of perpetual reverberant neural activity, even in the absence of specific tasks or stimuli. Shedding light on the origin and functional significance of such a dynamical state is essential to understanding how the brain transmits, processes, and stores information. An inspiring, albeit controversial, conjecture proposes that some statistical characteristics of empirically observed neuronal activity can be understood by assuming that brain networks operate in a dynamical regime near the edge of a phase transition. Moreover, the resulting critical behavior, with its concomitant scale invariance, is assumed to carry crucial functional advantages. Here, we present a data-driven analysis based on simultaneous high-throughput recordings of the activity of thousands of individual neurons in various regions of the mouse brain. To analyze these data, we synergistically combine cutting-edge methods for the study of brain activity (such as a phenomenological renormalization group approach and techniques that infer the general dynamical state of a neural population), while designing complementary tools. This strategy allows us to uncover strong signatures of scale invariance that is "quasi-universal" across brain regions and reveal that all these areas operate, to a greater or lesser extent, near the edge of instability. Furthermore, this framework allows us to distinguish between quasi-universal background activity and non-universal input-related activity. Taken together, this study provides strong evidence that brain networks actually operate in a critical regime which, among other functional advantages, provides them with a scale-invariant substrate of activity covariances that can sustain optimal input representations., Comment: 22 pages, 5 figures

Shedding light onto how biological systems represent, process and store information in noisy environments is a key and challenging goal. A stimulating, though controversial, hypothesis poses that operating in dynamical regimes near the edge of a phase transition, i.e. at criticality or the "edge of chaos", can provide information-processing living systems with important operational advantages, creating, e.g., an optimal trade-off between robustness and flexibility. Here, we elaborate on a recent theoretical result, which establishes that the spectrum of covariance matrices of neural networks representing complex inputs in a robust way needs to decay as a power-law of the rank, with an exponent close to unity, a result that has been indeed experimentally verified in neurons of the mouse visual cortex. Aimed at understanding and mimicking these results, we construct an artificial neural network and train it to classify images. Remarkably, we find that the best performance in such a task is obtained when the network operates near the critical point, at which the eigenspectrum of the covariance matrix follows the very same statistics as actual neurons do. Thus, we conclude that operating near criticality can also have -- besides the usually alleged virtues -- the advantage of allowing for flexible, robust and efficient input representations.

Reservoir Computing (RC) is an appealing approach in Machine Learning that combines the high computational capabilities of Recurrent Neural Networks with a fast and easy training method. Likewise, successful implementation of neuro-inspired plasticity rules into RC artificial networks has boosted the performance of the original models. In this manuscript, we analyze the role that plasticity rules play on the changes that lead to a better performance of RC. To this end, we implement synaptic and non-synaptic plasticity rules in a paradigmatic example of RC model: the Echo State Network. Testing on nonlinear time series prediction tasks, we show evidence that improved performance in all plastic models are linked to a decrease of the pair-wise correlations in the reservoir, as well as a significant increase of individual neurons ability to separate similar inputs in their activity space. Here we provide new insights on this observed improvement through the study of different stages on the plastic learning. From the perspective of the reservoir dynamics, optimal performance is found to occur close to the so-called edge of instability. Our results also show that it is possible to combine different forms of plasticity (namely synaptic and non-synaptic rules) to further improve the performance on prediction tasks, obtaining better results than those achieved with single-plasticity models.

It has been recently discovered that the measles virus can wipe out the adaptive immune system, destroying B lymphocytes and reducing the diversity of non-specific B cells of the infected host. In particular, this implies that previously acquired immunization from vaccination or direct exposition to other pathogens could be erased in a phenomenon named "immune amnesia", whose effects can become particularly worrisome given the actual rise of anti-vaccination movements. Here we present the first attempt to incorporate immune amnesia into standard models of epidemic spreading. In particular, we analyze diverse variants of a model that describes the spreading of two concurrent pathogens causing measles and another generic disease: the SIR-IA model. Analytical and computational studies confirm that immune amnesia can indeed have important consequences for epidemic spreading, significantly altering the vaccination coverage required to reach herd-immunity for concurring infectious diseases. More specifically, we uncover the existence of novel propagating and endemic phases which are induced by immune amnesia, that appear both in fully-connected and more structured networks, such as random networks and power-law degree-distributed ones. In particular, the transitions from a quiescent state into these novel phases can become rather abrupt in some cases that we specifically analyze. Furthermore, we discuss the meaning and consequences of our results and their relation with, e.g., immunization strategies, together with the possibility that explosive types of transitions may emerge, making immune-amnesia effects particularly dramatic. This work opens the door to further developments and analyses of immune amnesia effects, contributing, more generally, to the theory of interacting epidemics on complex networks.