LINE Solver Files

"Natural" sorting is resource intensive. Large inputs are slow to process. Use "Alphabetical" sort for large inputs. "Sort by delimiter" can sort by first, second ... last character, first, second ... last word in "Alphabetical" and "Natural" mode only. chore: added matlab-avro dependency chore: improve screen diagnostics for EnsembleSolver Click "Alphabetical" for case insensitive, ascending, alphabetical sorting. Click attached "Cs" button for case sensitive results. Click "Length" to sort lines by character length. Pre-sort alphabetically for length sorted results to be in alphabetical order. Click "Natural" for case insensitive natural sorting of lines. Click attached "Cs" button for case sensitive results. Click "Random" to randomize line order. Click "Reverse" to reverse current line order. Use to create descending order from ascending sorted lists. Default sorting column will be the first character of each new linfeat: added getStruct method to LayeredNetworks docs: updated manual and wiki with tget and ResidT field Enter your text/list for sorting here. feat: added ca and recal bigint implementations in NC feat: added calculation of residence times (ResidT) in getAvg and getAvgTable feat: added customizable attribute property to Model and Network element classs feat: added Geometric discrete distribution feat: added getInverseSubgenerator to MarkovianDistribution feat: added LINE version inside Model object feat: added mom and comom algorithms to NC feat: added proportionally fair estimator in NC feat: added sampling and adaptive methods in NC feat: added table get (tget) method feat: enableCheck property to disable support() checks in Solvers feat: entirely new LN solver feat: improved KT method in NC feat: NC solver now more accurate feat: new getStruct method in LayeredNetwork objects feat: scheduling type now categorical instead than string feat: support for asynchronous requests in LN fix: added numerical tolerance to Distrib.isImmediate fix: bug in definition of qn.proc for Immediate transitions fix: changed default behaviour of fromMarginalAndStarted, now force option is true by default fix: changed reset behaviour in linkAndLog.m to avoid deleting model metrics fix: explicit error thrown for finite multiplicities in LQN tasks with inf scheduling fix: sink classes misconfigured when adding node after closed class definitions Note: For files too large to successfully load into this text field, use the "Big File Tool" link below. perf: added caching to refreshEnsemble perf: added method to disable Network object checks perf: improved efficiency of Network.link method perf: improved routing matrix generation perf: improvements to Distrib methods perf: improvements to Network methods perf: NC default sampling now 1e5 samples perf: NC faster if pfqn_nc returns throughput and queue-lengths in a single iteration Privacy of Data: This tool is built-with and functions-in Client Side JavaScripting, so only your computer will see or process your data input/output. refactor: Class entry in all AvgTable renamed into JobClass refactor: Distrib tolerances changed to 1e-8/1e8 to match Java Integer range for mom/comom refactor: EnsembleSolver.run() renamed into iterate() refactor: getGenerator methods in MarkovianDistribution renamed to getSubgenerator refactor: LayeredNetwork Processors renamed as Host but backward compatible refactor: line version information now hidden refactor: NetworkSolverLibrary to Library refactor: re-enabled parfor option in EnsembleSolver refactor: removed Semantics class, stage annotation in random environments is now a custom string refactor: round robin policy now called RRB refactor: Solver.run() renamed into Solver.runAnalysis()e. Default setting will contain an empty delimiter and column number of 1. To sort via the second character enter a column number of 2. Sort via second word by entering a space as delimiter and column number of 2.