Julie Mullen
YOU?
Author Swipe
View article: Easy Acceleration with Distributed Arrays
Easy Acceleration with Distributed Arrays Open
High level programming languages and GPU accelerators are powerful enablers for a wide range of applications. Achieving scalable vertical (within a compute node), horizontal (across compute nodes), and temporal (over different generations …
View article: DBOS Network Sensing: A Web Services Approach to Collaborative Awareness
DBOS Network Sensing: A Web Services Approach to Collaborative Awareness Open
DBOS (DataBase Operating System) is a novel capability that integrates web services, operating system functions, and database features to significantly reduce web-deployment effort while increasing resilience. Integration of high performan…
View article: GPU Sharing with Triples Mode
GPU Sharing with Triples Mode Open
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. …
View article: LLload: An Easy-to-Use HPC Utilization Tool
LLload: An Easy-to-Use HPC Utilization Tool Open
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 Supercomputin…
View article: HPC with Enhanced User Separation
HPC with Enhanced User Separation Open
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 explo…
View article: Anonymized Network Sensing Graph Challenge
Anonymized Network Sensing Graph Challenge Open
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 th…
View article: What is Normal? A Big Data Observational Science Model of Anonymized Internet Traffic
What is Normal? A Big Data Observational Science Model of Anonymized Internet Traffic Open
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…
View article: Mapping of Internet "Coastlines" via Large Scale Anonymized Network Source Correlations
Mapping of Internet "Coastlines" via Large Scale Anonymized Network Source Correlations Open
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 sc…
View article: pPython Performance Study
pPython Performance Study Open
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 (Pyt…
View article: Focusing and Calibration of Large Scale Network Sensors using GraphBLAS Anonymized Hypersparse Matrices
Focusing and Calibration of Large Scale Network Sensors using GraphBLAS Anonymized Hypersparse Matrices Open
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 re…
View article: Deployment of Real-Time Network Traffic Analysis using GraphBLAS Hypersparse Matrices and D4M Associative Arrays
Deployment of Real-Time Network Traffic Analysis using GraphBLAS Hypersparse Matrices and D4M Associative Arrays Open
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.or…
View article: Hypersparse Network Flow Analysis of Packets with GraphBLAS
Hypersparse Network Flow Analysis of Packets with GraphBLAS Open
Internet analysis is a major challenge due to the volume and rate of network traffic. In lieu of analyzing traffic as raw packets, network analysts often rely on compressed network flows (netflows) that contain the start time, stop time, s…
View article: Python Implementation of the Dynamic Distributed Dimensional Data Model
Python Implementation of the Dynamic Distributed Dimensional Data Model Open
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 h…
View article: pPython for Parallel Python Programming
pPython for Parallel Python Programming Open
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 (Pyt…
View article: Temporal Correlation of Internet Observatories and Outposts
Temporal Correlation of Internet Observatories and Outposts Open
The Internet has become a critical component of modern civilization requiring scientific exploration akin to endeavors to understand the land, sea, air, and space environments. Understanding the baseline statistical distributions of traffi…
View article: GraphBLAS on the Edge: Anonymized High Performance Streaming of Network Traffic
GraphBLAS on the Edge: Anonymized High Performance Streaming of Network Traffic Open
Long range detection is a cornerstone of defense in many operating domains (land, sea, undersea, air, space, ..,). In the cyber domain, long range detection requires the analysis of significant network traffic from a variety of observatori…
View article: Spatial Temporal Analysis of 40,000,000,000,000 Internet Darkspace Packets
Spatial Temporal Analysis of 40,000,000,000,000 Internet Darkspace Packets Open
The Internet has never been more important to our society, and understanding\nthe behavior of the Internet is essential. The Center for Applied Internet Data\nAnalysis (CAIDA) Telescope observes a continuous stream of packets from an\nunso…
View article: 3D Real-Time Supercomputer Monitoring
3D Real-Time Supercomputer Monitoring Open
Supercomputers are complex systems producing vast quantities of performance data from multiple sources and of varying types. Performance data from each of the thousands of nodes in a supercomputer tracks multiple forms of storage, memory, …
View article: Vertical, Temporal, and Horizontal Scaling of Hierarchical Hypersparse GraphBLAS Matrices
Vertical, Temporal, and Horizontal Scaling of Hierarchical Hypersparse GraphBLAS Matrices Open
Hypersparse matrices are a powerful enabler for a variety of network, health,\nfinance, and social applications. Hierarchical hypersparse GraphBLAS matrices\nenable rapid streaming updates while preserving algebraic analytic power and\ncon…
View article: Supercomputing Enabled Deployable Analytics for Disaster Response
Supercomputing Enabled Deployable Analytics for Disaster Response Open
First responders and other forward deployed essential workers can benefit\nfrom advanced analytics. Limited network access and software security\nrequirements prevent the usage of standard cloud based microservice analytic\nplatforms that …
View article: Node-Based Job Scheduling for Large Scale Simulations of Short Running Jobs
Node-Based Job Scheduling for Large Scale Simulations of Short Running Jobs Open
Diverse workloads such as interactive supercomputing, big data analysis, and large-scale AI algorithm development, requires a high-performance scheduler. This paper presents a novel node-based scheduling approach for large scale simulation…
View article: Fast Mapping onto Census Blocks
Fast Mapping onto Census Blocks Open
Pandemic measures such as social distancing and contact tracing can be\nenhanced by rapidly integrating dynamic location data and demographic data.\nProjecting billions of longitude and latitude locations onto hundreds of\nthousands of hig…
View article: Large Scale Parallelization Using File-Based Communications
Large Scale Parallelization Using File-Based Communications Open
In this paper, we present a novel and new file-based communication architecture using the local filesystem for large scale parallelization. This new approach eliminates the issues with filesystem overload and resource contention when using…
View article: A Billion Updates per Second Using 30,000 Hierarchical In-Memory D4M Databases
A Billion Updates per Second Using 30,000 Hierarchical In-Memory D4M Databases Open
Analyzing large scale networks requires high performance streaming updates of graph representations of these data. Associative arrays are mathematical objects combining properties of spreadsheets, databases, matrices, and graphs, and are w…
View article: Hyperscaling Internet Graph Analysis with D4M on the MIT SuperCloud
Hyperscaling Internet Graph Analysis with D4M on the MIT SuperCloud Open
Detecting anomalous behavior in network traffic is a major challenge due to the volume and velocity of network traffic. For example, a 10 Gigabit Ethernet connection can generate over 50 MB/s of packet headers. For global network providers…
View article: Scalability of VM provisioning systems
Scalability of VM provisioning systems Open
Virtual machines and virtualized hardware have been around for over half a\ncentury. The commoditization of the x86 platform and its rapidly growing\nhardware capabilities have led to recent exponential growth in the use of\nvirtualization…
View article: Enhancing HPC security with a user-based firewall
Enhancing HPC security with a user-based firewall Open
HPC systems traditionally allow their users unrestricted use of their\ninternal network. While this network is normally controlled enough to guarantee\nprivacy without the need for encryption, it does not provide a method to\nauthenticate …
View article: Benchmarking SciDB data import on HPC systems
Benchmarking SciDB data import on HPC systems Open
SciDB is a scalable, computational database management system that uses an\narray model for data storage. The array data model of SciDB makes it ideally\nsuited for storing and managing large amounts of imaging data. SciDB is\ndesigned to …