JuliaHub Blog: Insights & Updates

Newsletter August 2022: Key Highlights | JuliaHub

Written by Andrew Claster | Aug 22, 2022

JuliaCon 2022 Highlights

This year’s JuliaCon was the biggest and best ever with more than 50 thousand unique viewers on YouTube and other platforms. All 229 presentations, workshops and other videos are available on YouTube.

This year’s highlights include:

And congratulations to the winners of the 2022 Julia Community Prizes!

  • Dr. Morten Piibeleht, Michael Hatherly, Dr. Fredrik Ekre, and Dr. Mauro Werder for their work on Documenter.jl and its ecosystem

  • Dr. Frames White for her many technical and community contributions across the Julia ecosystem

  • Shuhei Kadowaki for his work on JET.jl and the Julia compiler

For more highlights, see JuliaCon 2022 Highlights from the JuliaCon 2022 Organizing Committee.

Pumas: Pumas is a comprehensive platform for pharmaceutical modeling and simulation in Julia. Click here for more information.

PumasQSP v2.0: PumasQSP is a comprehensive cloud-based platform for quantitative systems pharmacological analytics. PumasQSP v2.0 is now available with new and improved features:

  • Bayesian inference is now available when generating virtual populations

  • The timespan of a given trial can be fit to data, similarly to how model parameters and initial conditions are optimized to match the data of the trial.

  • Subsampling a virtual population of patients vp for a given trial can now be performed with a simple call to subsample(alg, vp, trial), where alg is one of several available subsampling methods, like MAPEL or Allen-Rieger-Musante 2016 ARM.

  • Users can now set target distributions for any number of model states and our new custom-made method will subsample a virtual population, so that the same model states of virtual patients match the target distributions, as measured by discretized histograms.

JuliaHub v5.7.1: JuliaHub v5.7.1 is now available. JuliaHub is the entry point for all things Julia: explore the ecosystem, build packages and deploy a supercomputer at the click of a button. JuliaHub also allows you to develop Julia applications interactively using a browser-based IDE or by using the Pluto notebook environment and then scale workloads to thousands of cores. Version 5 features a brand new user interface, reduced app startup latency, and many more usability enhancements. JuliaHub is the easiest way to start developing in Julia or share your work using dashboards and notebooks. JuliaHub v5.7.1 contains the following new features:

  • CPU and Memory usage of jobs are now available as charts

  • Bug fixes

  • Enterprise

    • Offline package server

    • New application: Windows Desktop

    • New feature: Project (Experimental)

JuliaSim v0.22.0: JuliaSim is a next generation cloud-based simulation platform, combining the latest techniques in SciML with equation-based digital twin modeling and simulation. JuliaSim v0.22.0 is now available with the following features:

  • The Functional Mockup Unit (FMU) Accelerator Graphical User Interface (GUI) provides a browser-based no-code workflow for generating surrogates of FMUs. It includes support for linear and non-linear continuous-time echo stem network (CTESN) and corresponding configuration options (i.e. reservoir size, number of sample points and time span to simulate over)

  • JuliaSimSurrogates.jl supports parallelized generation of simulation data that scales with available compute for ModelingToolKit.jl models and FMUs for Model Exchange/CoSimulation.

  • JuliaSimControls.jl includes Autotuning of proportional-integral (PI) / proportional-integral-<wbr>derivative (PID) controllers and a Pluto-based GUI for autotuning, with:

    • Linear constrained model predictive control (MPC) functionality

    • Nonlinear MPC functionality using sequential quadratic programming (SQP)

    • A library of state observers, including Kalman filter, Extended and Unscented Kalman filter (EKF/UKF)

    • Linear analysis of ModelingToolkit models through frequency-response analysis

    • Tuning-objective types with plot recipes

    • Inverse-optimal control for robust MPC tuning

    • Linear model reduction and controller-order reduction

    • Pluto-based GUI for model reduction

    • Full interoperability with ControlSystems.jl and RobustAndOptimalControl.jl

  • DFTSurrogates.jl provides a set of surrogates for computational chemistry:

    • Reading and writing of crystallographic interchange format (CIF) / XYZ / simplified molecular-input line-entry system (SMILES) format

    • Compatibility with MolecularGraph.jl for graph representation, Xtals.jl for representing crystals and AtomsBase.jl for generic atomic properties

    • Generating featurization pipelines based on atomic, pair and bond properties

    • Surrogates by Crystal Graph Convolutional Neural Networks

    • Parallelization of inference for multiple inputs candidate molecules

    • Ability to train surrogates with custom architectures and datasets

    • Interoperability with Chemellia

Mathematical Programming with Julia: Mathematical Programming with Julia is a new book by Richard Lusby and Thomas Stidsen. It features an open source approach to linear and mixed-integer programming. Click here for more.

Julia for Election Forecasting: Do you know a media company interested in covering the intersection of politics and statistics? The TuringElect team is looking for media partners interested in Bayesian forecasts for the upcoming midterms! More information is available here, and you can also find a sneak preview of what their model can do. Contact cdp49@cam.ac.uk for more details.

Julia for Power Dynamics: Continuous-Time Echo State Networks for Predicting Power System Dynamics is a new paper co-authored by Dr. Chris Rackauckas, Julia Computing Director of Modeling and Simulation. “Continuous-time echo state networks (with hybrid terms) can accurately predict the (highly stiff) dynamics of power system dynamics,” Chris explains. The paper is based on work with Lawrence Livermore National Laboratory on surrogates of differential-algebraic equation (DAE) systems. Click here for more.

Julia for Power Grid Optimization: Society for Industrial and Applied Mathematics (SIAM) News has published Rapid Prototyping with Julia: From Mathematics to Fast Code which explains how Julia is used for power grid optimization with ExaSGD as part of the US Department of Energy’s Exascale Computing Project (ECP). Click here for more.

Julia for Market Prediction: “G-Research is looking for a software engineer who is keen to contribute directly to the open-source Julia project. G-Research uses data science and machine learning tools to predict movements in the markets and we're very interested in furthering the development of the Julia language and supporting the community. This role could focus on a number of different areas from the compiler to Flux to improved packaging to DataFrames – there's a lot to do and we're looking for someone who has a passion to move an area of Julia forward. This role will be a part of our open-source program office so all contributions from this role will definitely impact the entire community. If you're interested, reach out at jobs@gr-oss.io"

Julia Computing - Coming to a Conference Near You: Julia Computing will be present at a number of upcoming conferences and events. Click below for more information.

Converting from Proprietary Software to Julia: Are you looking to leverage Julia’s superior speed and ease of use, but limited due to legacy software and code? Julia Computing and our partners can help accelerate replacing your existing proprietary applications, improve performance, reduce development time, augment or replace existing systems and provide an extended trusted team to deliver Julia solutions. Leverage experienced resources from Julia Computing and our partners to get your team up and running quickly. For more information, please contact us.

Careers at Julia Computing: Julia Computing is a fast-growing tech company with fully remote employees in 12 countries on 5 continents. Click here to learn more about exciting careers and internships with Julia Computing.

Julia and Julia Computing in the News

Julia Blog Posts

Recent Julia Online Events

Contact Us: Please contact us if you wish to:

  • Purchase or obtain license information for products such as JuliaHub, JuliaSim or Pumas

  • Obtain pricing for Julia consulting projects for your organization

  • Schedule online Julia training for your organization

  • Share information about exciting new Julia case studies or use cases

  • Spread the word about an upcoming online event involving Julia

  • Partner with Julia Computing to organize a Julia event online

  • Submit a Julia internship, fellowship or job posting

About Julia Computing and Julia

Julia Computing's mission is to develop products that bring Julia's superpowers to its customers. Julia Computing's flagship product is JuliaHub, a secure, software-as-a-service platform for developing Julia programs, deploying them, and scaling to thousands of nodes. It provides the power of a supercomputer at the fingertips of every data scientist and engineer. In addition to data science workflows, JuliaHub also provides access to cutting-edge products such as Pumas for pharmaceutical modeling and simulation, JuliaSim for multi-physics modeling and simulation, and Cedar for electronic circuit simulation, combining traditional simulation with modern SciML approaches.

Julia is the fastest high performance open source computing language for data, analytics, algorithmic trading, machine learning, artificial intelligence, and other scientific and numeric computing applications. Julia solves the two language problem by combining the ease of use of Python and R with the speed of C++. Julia provides parallel computing capabilities out of the box and unlimited scalability with minimal effort. Julia has been downloaded by users at more than 10,000 companies and is used at more than 1,500 universities. Julia co-creators are the winners of the 2019 James H. Wilkinson Prize for Numerical Software and the 2019 Sidney Fernbach Award. Julia has run at petascale on 650,000 cores with 1.3 million threads to analyze over 56 terabytes of data using Cori, one of the ten largest and most powerful supercomputers in the world.
View full post