Crew Launcher Plugins for Traditional High-Performance Computing Clusters

In computationally demanding analysis projects, statisticians and data scientists asynchronously deploy long-running tasks to distributed systems, ranging from traditional clusters to cloud services. The 'crew.cluster' package extends the 'mirai'-powered 'crew' package with worker launcher plugins for traditional high-performance computing systems. Inspiration also comes from packages 'mirai' by Gao (2023) < https://github.com/r-lib/mirai>, 'future' by Bengtsson (2021) , 'rrq' by FitzJohn and Ashton (2023) < https://github.com/mrc-ide/rrq>, 'clustermq' by Schubert (2019) ), and 'batchtools' by Lang, Bischl, and Surmann (2017). .

A Distributed Worker Launcher Framework

In computationally demanding analysis projects, statisticians and data scientists asynchronously deploy long-running tasks to distributed systems, ranging from traditional clusters to cloud services. The 'NNG'-powered 'mirai' R package by Gao (2023) is a sleek and sophisticated scheduler that efficiently processes these intense workloads. The 'crew' package extends 'mirai' with a unifying interface for third-party worker launchers. Inspiration also comes from packages. 'future' by Bengtsson (2021) , 'rrq' by FitzJohn and Ashton (2023) < https://github.com/mrc-ide/rrq>, 'clustermq' by Schubert (2019) ), and 'batchtools' by Lang, Bischel, and Surmann (2017) .

Robust, Logged and Reproducible Iteration at Organizational Scale

Turns one-off iterative R procedures (such as for loops, lapply() or pmap() from 'purrr') into production-grade workflows by wrapping them with orthogonal, composable execution layers. Two layers are always active: structured logging with real traceback and per-case timing; and reproducibility capture, which records the R version, loaded package versions, execution environment, the exact iteration mask, and a stat-based fingerprint of every input file referenced in the mask (with a diff_inputs() helper to detect silent drift between runs). Parallel execution (built on the 'future' framework, Bengtsson (2021) ), non-blocking background jobs, and opt-in progress reporting (via 'progressr') are implemented as optional, composable layers. Further layers (error replay, content-hash input fingerprinting, content-based case identifiers) are planned and will remain composable with the default layers.

Spatial Parallel Computing by Hierarchical Data Partitioning

Geospatial data computation is parallelized by grid, hierarchy, or raster files. Based on 'future' (Bengtsson, 2024 ) and 'mirai' (Gao et al., 2025 ) parallel back-ends, 'terra' (Hijmans et al., 2025 ) and 'sf' (Pebesma et al., 2024 ) functions as well as convenience functions in the package can be distributed over multiple threads. The simplest way of parallelizing generic geospatial computation is to start from par_pad_*() functions to par_grid(), par_hierarchy(), or par_multirasters() functions. Virtually any functions accepting classes in 'terra' or 'sf' packages can be used in the three parallelization functions. A common raster-vector overlay operation is provided as a function extract_at(), which uses 'exactextractr' (Baston, 2023 ), with options for kernel weights for summarizing raster values at vector geometries. Other convenience functions for vector-vector operations including simple areal interpolation (summarize_aw()) and summation of exponentially decaying weights (summarize_sedc()) are also provided.

Latent Variable Models Diagnostics

Diagnostics and visualization tools for latent variable models fitted with 'lavaan' (Rosseel, 2012 ). The package provides fast, parallel-safe factor-score prediction (lavPredict_parallel()), data augmentation with model predictions, residuals, delta-method standard errors and confidence intervals (augment()), and model-based latent grids for continuous, ordinal, or mixed indicators (prepare()). It offers item-level empirical versus model curve comparison using generalized additive models for both continuous and ordinal indicators (item_data(), item_plot()) via 'mgcv' (Wood, 2017, ISBN:9781498728331), residual diagnostics including residual correlation tables and plots (resid_cor(), resid_corrplot()) using 'corrplot' (Wei and Simko, 2021 < https://github.com/taiyun/corrplot>), and Q–Q checks of residual z-statistics (resid_qq()), optionally with non-overlapping labels from 'ggrepel' (Slowikowski, 2024 < https://CRAN.R-project.org/package=ggrepel>). Heavy computations are parallelized via 'future'/'furrr' (Bengtsson, 2021 ; Vaughan and Dancho, 2018 < https://CRAN.R-project.org/package=furrr>). Methods build on established literature and packages listed above.

Genetic Population Level Functions

This collection of gene representation-independent functions implements the population layer of extended evolutionary and genetic algorithms and its support for the R-package 'xega' < https://CRAN.R-project.org/package=xega>. The population layer consists of functions for initializing, logging, observing, evaluating a population of genes, as well as of computing the next population. For parallel evaluation of a population of genes 4 execution models - named Sequential, MultiCore, FutureApply, and Cluster - are provided. They are implemented by configuring the lapply() function. The execution model FutureApply can be externally configured as recommended by Bengtsson (2021) . Configurable acceptance rules and cooling schedules (see Kirkpatrick, S., Gelatt, C. D. J, and Vecchi, M. P. (1983) , and Aarts, E., and Korst, J. (1989, ISBN:0-471-92146-7) offer simulated annealing or greedy randomized approximate search procedure elements. Adaptive crossover and mutation rates depending on population statistics generalize the approach of Stanhope, S. A. and Daida, J. M. (1996, ISBN:0-18-201-031-7). For 'xega''s architecture, see Geyer-Schulz, A. (2025) .

Understand and Describe Bayesian Models and Posterior Distributions

Provides utilities to describe posterior distributions and Bayesian models. It includes point-estimates such as Maximum A Posteriori (MAP), measures of dispersion (Highest Density Interval - HDI; Kruschke, 2015 ) and indices used for null-hypothesis testing (such as ROPE percentage, pd and Bayes factors). References: Makowski et al. (2021) .

R Fortunes

A collection of fortunes from the R community.