Utilities and Constants

Constants and enumerations

These modules provide utility functions, constants, and enumerations.

Constants and Enumerations (matlab/src/lang/constant/)

The constants module defines enumerations and global constants used throughout the LINE Solver package.

Scheduling Strategies (SchedStrategy)

FCFS: First-Come First-Served
LCFS: Last-Come First-Served (non-preemptive)
LCFSPR: Last-Come First-Served Preemptive-Resume
LCFSPI: Last-Come First-Served Preemptive-Independent
PS: Processor Sharing (egalitarian)
DPS: Discriminatory Processor Sharing
GPS: Generalized Processor Sharing
INF: Infinite Server (delay station)
HOL: Head-of-Line priority (non-preemptive)
SJF: Shortest Job First
LJF: Longest Job First
SEPT: Shortest Expected Processing Time
LEPT: Longest Expected Processing Time
SIRO: Service In Random Order
POLLING: Polling-based service
EXT: External (arrivals from outside)
FORK: Fork behavior
REF: Reference (for reference stations)

Routing Strategies (RoutingStrategy)

DISABLED: Routing disabled
FIRING: Petri net transition firing
JSQ: Join Shortest Queue
KCHOICES: Power of k choices
PROB: Probabilistic routing
RAND: Random (uniform) routing
RROBIN: Round-robin routing
WRROBIN: Weighted round-robin

Process Types (ProcessType)

EXP: Exponential
ERLANG: Erlang
HYPEREXP: Hyperexponential
APH: Acyclic Phase-Type
PH: Phase-Type
COXIAN: Coxian
DET: Deterministic
UNIFORM: Uniform
GAMMA: Gamma
PARETO: Pareto
WEIBULL: Weibull
LOGNORMAL: Log-normal
MAP: Markovian Arrival Process
MMPP2: Two-state MMPP
REPLAYER: Trace replayer
TRACE: Generic trace
DISABLED: Disabled process
IMMEDIATE: Instantaneous

Node Types (NodeType)

SOURCE: Arrival source
SINK: Departure sink
QUEUE: Service queue
DELAY: Infinite server station
ROUTER: Routing node
FORK: Fork node
JOIN: Join node
CACHE: Cache station
CLASSSWITCH: Class-switching node
LOGGER: Logging node
PLACE: Petri net place
TRANSITION: Petri net transition

Job Class Types (JobClassType)

OPEN: Open class (external arrivals)
CLOSED: Closed class (fixed population)
DISABLED: Disabled class

Service Strategies (ServiceStrategy)

LI: Load-Independent service
LD: Load-Dependent service

Solver Types (SolverType)

MVA: Mean Value Analysis
CTMC: Continuous-Time Markov Chain
FLUID: Fluid approximation
MAM: Matrix-Analytic Methods
NC: Normalizing Constant
SSA: Stochastic Simulation Algorithm
JMT: Java Modelling Tools
AUTO: Automatic selection
LQNS: Layered QN Solver
QNS: QN Solver
LN: Layered Network
ENV: Random Environment

Metric Types (MetricType)

ArvR: Arrival rate
DropRate: Drop rate
FCRMemOcc: FCR memory occupation
FCRWeight: FCR total weight
FJQLen: Fork-join number of customers
FJRespT: Fork-join response time
QLen: Number of customers (queue length)
QueueT: Queue time
ResidT: Residence time
RespT: Response time
RespTSink: Response time per sink
SysQLen: System number of customers
SysRespT: System response time
SysTput: System throughput
Tput: Throughput
TputSink: Throughput per sink
TranQLen: Transient number of customers
TranRespT: Transient response time
TranTput: Transient throughput
TranUtil: Transient utilization
Util: Utilization

Drop Strategies (DropStrategy)

WAITQ: Waiting queue (infinite buffer)
DROP: Drop arriving customers when full
BAS: Blocking after service

Replacement Strategies (ReplacementStrategy)

LRU: Least Recently Used
FIFO: First-In First-Out
RR: Random Replacement
SFIFO: Static FIFO
MRU: Most Recently Used
LFU: Least Frequently Used

Join Strategies (JoinStrategy)

STD: Standard join (wait for all)
PARTIAL: Partial join (k-of-n)
GUARD: Guarded join

Call Types (CallType)

SYNC: Synchronous call
ASYNC: Asynchronous call

Activity Precedence Types (ActivityPrecedenceType)

PRE_SEQ: Sequential precedence
PRE_AND: AND split/join
PRE_OR: OR split/join
POST_SEQ: Post-sequential
POST_AND: Post-AND
POST_OR: Post-OR
POST_LOOP: Post-loop

Polling Types (PollingType)

EXHAUSTIVE: Exhaustive polling
GATED: Gated polling
KLIMITED: K-limited polling

Verbose Levels (VerboseLevel)

SILENT: No output
STD: Standard output
DEBUG: Debug output

GlobalConstants

The GlobalConstants class provides global configuration options:

CoarseTol: Coarse tolerance for comparisons (default: 1e-3)
FineTol: Fine tolerance for comparisons (default: 1e-6)
Zero: Near-zero threshold (default: 1e-14)
Immediate: Immediate service rate (default: Inf)
MaxInt: Maximum integer value (default: 2147483647)

Usage Examples

Setting Scheduling Strategy

model = Network('Example');
queue = Queue(model, 'Queue1', SchedStrategy.PS);  % Processor Sharing

Setting Routing Strategy

model.link(Network.serialRouting(source, queue, sink));
% or use probabilistic routing
P = model.initRoutingMatrix();
P{class1}(source, queue) = 1.0;
P{class1}(queue, sink) = 1.0;
model.link(P);

Setting Distribution Type

queue.setService(jobclass, Exp(1.0));       % Exponential
queue.setService(jobclass, Erlang(2, 0.5)); % Erlang
queue.setService(jobclass, HyperExp([0.5 0.5], [1.0 2.0])); % Hyperexponential