Modelo de benchmarks
dispatchatlas.bench define la evidencia de benchmarks antes de que los solvers o las
campañas la consuman. Una familia de benchmarks declara su taxonomía, perfil de dominio,
clase de perfil, supuestos, evidencia de citación, envolvente de escala, espacio de
nombres de semilla, y esquema de salida. La materialización valida cada problema generado
con dispatchatlas.core, caracteriza la instancia, la envuelve en un sobre de
procedencia, y registra hashes estables.
El catálogo cubre familias genéricas de planificación de optimización-combinatoria y planificación de distributed-computing como pares co-iguales, de modo que la plataforma no es una herramienta solo-de-distributed-computing.
Ejemplo ejecutable: examples/benchmark_continuum.py genera, caracteriza, y cataloga un atlas de benchmarks continuo al vuelo.
Familias de planificación
Cada familia de planificación se materializa como un par de catálogo de primera-clase con al menos un perfil generador. La tabla de catálogo de abajo se genera a partir del registro de generadores de benchmarks y la matriz de citación, de modo que sus totales de familia y citaciones de fuente son contables desde las propias filas. Un gráfico de distribución-de-familias encima de la tabla muestra cómo los perfiles de familia se reparten entre las categorías de familias de planificación.
Generated from the benchmark generator registry and the citation matrix: 69 family profiles across 8 scheduling families — distributed-computing (47), flow-shop (7), job-shop (6), machine-scheduling (3), open-shop (1), rcpsp (3), rcpsp-max (1), setup-flow-shop (1).
Showing 69 of 69 family profiles.
| Distinctive against | Evidence | Citation status | Source citations | |||
|---|---|---|---|---|---|---|
accelerator-coschedulingaccelerator-coscheduling a heterogeneous datacenter job needs a compute resource and a scarce accelerator at the same time, so every job co-allocates two simultaneous resource demands held together for its whole run; the accelerator pool is scarce, so jobs sharing an accelerator serialize on it while jobs on disjoint resources run in parallel, and the scheduler reasons over a multi-resource co-allocation problem rather than a single-unit-demand one | distributed-computing | ioe-complete | published single-resource continuum schedulers (one unit resource demand per task, not the simultaneous co-allocation of a compute resource and a scarce accelerator held together under a Pareto contract) | smoke | citation-backed |
|
aerial-edgeaerial-edge a loitering UAV is a flying fog node that serves the ground region beneath it for a fixed loiter window before moving to the next pass, so sorties group into successive loiter passes pinned to the fog tier the platform embodies while overhead | distributed-computing | ioe-complete | aerial-edge MEC simulators (no flying-fog loiter placement) | smoke | citation-backed |
|
anytime-inferenceanytime-inference an edge accelerator serves deep-learning inference requests that each complete a mandatory minimal-accuracy early-exit branch and may run an optional refinement to full accuracy when their latency deadline allows; every request declares a mandatory duration below its full duration and a tight latency deadline, and arrivals are spaced shorter than a full inference so requests queue and contend, so the scheduler decides which requests refine and which deliver the early-exit result -- an imprecise-computation quality-versus-timeliness trade-off | distributed-computing | ioe-complete | published edge-cloud split inference and datacenter inference serving (a fixed full-accuracy computation per request), neither of which lets a request drop an optional refinement to meet its deadline so the schedule order trades accuracy for timeliness | smoke | citation-backed |
|
blocking-flow-shopblocking-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
bulk-synchronous-graphbulk-synchronous-graph an iterative graph computation runs as a sequence of supersteps separated by global barriers, so each superstep's vertex partitions compute and exchange messages and every partition of the next superstep waits on all partitions of the prior one; the slowest partition therefore gates each superstep, and the scheduler reasons over a barrier-synchronized partition-balancing problem rather than an overlap-friendly pipeline or an independent-task one | distributed-computing | ioe-complete | published pipeline or independent-task schedulers (an overlap-friendly wavefront or unsynchronized tasks, not supersteps separated by global barriers where the slowest partition gates each round under a Pareto contract) | smoke | citation-backed |
|
carbon-awarecarbon-aware flexible jobs defer to low-carbon-intensity windows under a time-varying grid carbon signal while honoring their SLA deadlines | distributed-computing | ioe-complete | Electricity-Maps grid carbon-intensity and CityLearn carbon-aware community signals | smoke | citation-backed |
|
cloud-independentcloud-edge-independent | distributed-computing | domain-specific | — | smoke | citation-backed |
|
coflow-schedulingcoflow-scheduling a distributed-computing stage completes only when the last network transfer of its coflow lands, not the first, so a coflow's completion time is the maximum over its member flows; every coflow emits data-heavy flow tasks that place freely across the fabric plus a barrier task that depends on all of them, so the barrier gates the group and the coflow-completion-time is an all-or-nothing footprint | distributed-computing | ioe-complete | datacenter coflow schedulers (no continuum tier-placement barrier) | smoke | citation-backed |
|
compact-job-shopcompact-job-shop | job-shop | classical | — | smoke | citation-backed |
|
confidential-edgeconfidential-edge tasks are classified by data sensitivity -- a confidential task that touches protected data must execute inside the enclave-capable trusted tier so its data never leaves the trusted boundary, while a public task draws a free placement affinity across the fabric | distributed-computing | ioe-complete | edge enclave runtimes (no security-classified Pareto placement) | smoke | citation-backed |
|
cyber-physicalcyber-physical control cycles arrive periodically under a time-varying tariff | distributed-computing | ioe-complete | periodic hard-real-time task models and smart-grid demand-side scheduling formulations (no edge-fog-cloud tier placement under a Pareto contract) | smoke | citation-backed |
|
data-localitydata-locality a query over a large dataset is cheaper to run where the data already resides than to ship the data across the WAN fabric; each task's input lives on one tier (edge sensor logs, fog warm aggregates, or cloud cold archives) and the task pins to that tier so its heavy input never crosses the fabric, with the data tiers cycled so placement spans the whole edge-fog-cloud continuum | distributed-computing | ioe-complete | cluster locality schedulers (no continuum data-residency placement) | smoke | citation-backed |
|
datacenter-colocationdatacenter-colocation latency-sensitive service jobs and deferrable batch jobs share multi-tenant cells, so high-priority-band jobs claim capacity ahead of low-band jobs under contention | distributed-computing | ioe-complete | Google Borg ClusterData2019 priority-tiered cell traces | smoke | citation-backed |
|
digital-twin-syncdigital-twin-sync each physical asset periodically syncs its state to its fog/cloud twin and must finish within a freshness (Age-of-Information) window before the twin's state goes stale | distributed-computing | ioe-complete | digital-twin edge frameworks (no joint Pareto placement) | smoke | citation-backed |
|
disaggregated-memorydisaggregated-memory a CXL-pooled cloud platform backs each socket with a small local DRAM tier and a shared far-memory pool, and every VM draws a long-tailed memory working set, so a few memory-hungry tenants dominate a socket's local budget while far-memory access inflates a VM's runtime in proportion to the working set it spills to the pool; the scheduler reasons over local-versus-pool placement rather than core-count placement | distributed-computing | ioe-complete | published CXL memory-pooling and tiered-memory systems (socket-local page placement, no edge-fog-cloud tier scheduling under a Pareto contract) | smoke | citation-backed |
|
distributed-assembly-flow-shopdistributed-assembly-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
distributed-flexible-job-shopdistributed-flexible-job-shop | job-shop | structurally-complex | — | smoke | citation-backed |
|
distributed-permutation-flow-shopdistributed-permutation-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
distributed-training-gangdistributed-training-gang a GPU cluster runs synchronous data-parallel training jobs; each job is a gang of workers that must START TOGETHER on distinct accelerators (every all-reduce step synchronizes the workers), so a job cannot begin until enough accelerators are free simultaneously; the workers reuse the accelerator pool and jobs arrive over time, so jobs queue and the scheduler decides which job acquires a full simultaneously-free worker set first -- an all-or-nothing gang co-start, not an independent placement of each worker | distributed-computing | ioe-complete | published GPU-cluster and inference-serving families that place each task independently; none requires a whole job's worker set to co-start simultaneously on distinct accelerators, so no other family forbids a partial start -- the gang-scheduling all-or-nothing constraint under a Pareto contract | smoke | citation-backed |
|
distributed-transactiondistributed-transaction a partitioned database runs transactions that each acquire exclusive locks on a variable read/write set of data shards, so every transaction co-allocates a randomly drawn subset of shards held together for its whole run; two transactions whose shard sets intersect serialize while disjoint transactions commit in parallel, so the scheduler reasons over a variable-cardinality lock-conflict graph rather than a fixed two-resource hold | distributed-computing | ioe-complete | published replica-placement schedulers (a single shard pinned per task for locality, not a variable-cardinality exclusive lock set co-allocated per transaction forming a conflict graph under a Pareto contract) | smoke | citation-backed |
|
edge-offloadingedge-offloading-mec each task chooses between local edge execution and remote offload | distributed-computing | ioe-complete | iFogSim MEC offloading scenarios | smoke | citation-backed |
|
edge-placementedge-placement services place on edge servers near their user population and migrate as demand shifts across base-station coverage cells | distributed-computing | ioe-complete | EUA edge-user-allocation and Shanghai-Telecom base-station placement traces | smoke | citation-backed |
|
elastic-serverless-autoscaleelastic-serverless-autoscale a serverless platform serves function invocations on a shared worker pool, and every invocation is moldable: it may run on one, two, or four concurrent workers, where a wider allocation runs shorter by a sublinear speedup but spends more total worker-seconds; each request declares its execution modes and arrivals are spaced shorter than a base invocation so the pool is contended, so the scheduler picks each invocation's worker width -- a moldable latency-versus-cost choice rather than a fixed resource hold | distributed-computing | ioe-complete | published serverless cold-start and fixed-width co-allocation families (accelerator co-scheduling, fpga partitioning), each of which holds one fixed resource set per task; none lets a request choose among several worker-count modes so the schedule order trades latency for resource cost under a Pareto contract | smoke | citation-backed |
|
facility-assignmentfacility-assignment | machine-scheduling | classical | — | smoke | citation-backed |
|
failure-recoveryfailure-recovery a failed task re-places its checkpoint state to a surviving tier | distributed-computing | ioe-complete | Borg cluster failure-event traces | smoke | citation-backed |
|
federated-learningfederated-learning each training round selects a subset of heterogeneous, straggler-prone edge clients that train on non-IID local data, then a fog or cloud aggregator combines their updates | distributed-computing | ioe-complete | FedScale and Oort federated-learning device-participation benchmarks | smoke | citation-backed |
|
flexible-job-shopflexible-job-shop | job-shop | classical | — | smoke | citation-backed |
|
flow-shoppermutation-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
fpga-partitioningfpga-partitioning a multi-tenant reconfigurable FPGA hosts tenant kernels that each occupy a contiguous region of fabric tiles, so every kernel co-allocates a contiguous run of tiles held together for its whole residency; two kernels whose tile intervals overlap cannot be co-resident and serialize while kernels on disjoint tile spans run in parallel, so the scheduler reasons over an interval-overlap conflict graph rather than a fixed two-resource hold or a random-subset lock set | distributed-computing | ioe-complete | published accelerator co-scheduling (a fixed compute-plus-accelerator pair) and shard-lock transactions (a random subset of resources), neither of which constrains the co-allocated set to a spatially contiguous tile interval whose overlaps form an interval conflict graph under a Pareto contract | smoke | citation-backed |
|
frontierco-fjspfrontierco-fjsp | job-shop | classical | — | smoke | citation-backed |
|
generative-inference-servinggenerative-inference-serving a transformer inference replica batches autoregressive requests that each hold key-value-cache memory proportional to their token count for the whole decode, so a long-tailed sequence mix fragments a fixed cache budget and the scheduler reasons over memory-bound admission rather than GPU-count placement; every request's decode duration and KV-cache demand scale with its drawn token count | distributed-computing | ioe-complete | published single-replica generative-model serving systems (no edge-fog-cloud tier placement under a Pareto contract) | smoke | citation-backed |
|
gpu-mlgpu-ml training and inference jobs claim accelerators and gang-schedule replicas | distributed-computing | ioe-complete | Alibaba PAI, Philly, and Helios GPU-cluster traces | smoke | citation-backed |
|
hybrid-flow-shophybrid-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
immersive-xrimmersive-xr each extended-reality frame runs a latency-critical perception, render, and display pipeline placed across the device, edge, and cloud within a hard motion-to-photon deadline | distributed-computing | ioe-complete | ILLIXR extended-reality systems testbed | smoke | citation-backed |
|
intermittent-edgeintermittent-edge a batteryless sensor harvests ambient energy into a small buffer, runs until the buffer depletes, then sleeps to recharge; a job too large for one duty-cycle window is checkpointed at power loss and resumed in the next, so it is a precedence chain of edge-pinned per-window segments each bounded by the constant energy window | distributed-computing | ioe-complete | intermittent-computing runtimes (no continuum energy-window placement) | smoke | citation-backed |
|
iot-edgeiot-edge many small sensor readings arrive periodically and aggregate at the edge | distributed-computing | ioe-complete | published wireless-sensor-network telemetry datasets and in-network aggregation deployments (raw sensor readings, not edge-fog-cloud tier scheduling under a Pareto contract) | smoke | citation-backed |
|
job-shopjob-shop | job-shop | classical | — | smoke | citation-backed |
|
kv-cache-placementkv-cache-placement a distributed key-value cache tier serves a catalog of cache objects whose request rate follows a heavy Zipfian popularity skew, so a few hot objects absorb most of the traffic; each object is placement-flexible, carrying one single-node mode per cache-tier node, so the scheduler chooses which tier node hosts it, and an object's working set -- the transfer volume staged onto its host tier -- scales with its popularity rank, so the hottest object carries the largest working set and a cold-tail object the smallest; a naive uniform placement strands a hot, large-working-set object on a far tier and pays its transfer across the fabric, while a locality-aware placement pins the hottest objects to near tiers to shrink makespan and cost | distributed-computing | ioe-complete | published cache and content-placement families (consistent-hashing replica placement, CDN content distribution) that place each object uniformly or by a hash; none scales each object's working set with a Zipfian popularity rank so the hot objects carry a strictly larger transfer volume, making popularity-skewed near-tier pinning the lever a locality-aware placement pulls under the Pareto contract | smoke | citation-backed |
|
machine-schedulingmachine-scheduling-unrelated | machine-scheduling | classical | — | smoke | citation-backed |
|
microservice-dagmicroservice-dag services form an acyclic call graph pinned by role to a tier | distributed-computing | ioe-complete | Alibaba v2021 microservice-trace call graphs | smoke | citation-backed |
|
mixed-criticalitymixed-criticality a safety-critical real-time mix runs tasks of differing criticality, and a high-criticality task is budgeted with a conservative high-assurance worst-case execution time and a tight deadline while a low-criticality task carries a smaller best-effort budget and a loose deadline, so the criticality tiering lives in the duration and deadline structure; the scheduler reasons over which assured-criticality tasks to guarantee under contention rather than a uniform-assurance deadline-scheduling one | distributed-computing | ioe-complete | published uniform-assurance real-time deadline schedulers (one worst-case execution time and deadline class per task, not criticality-tiered WCET budgets with tighter high-assurance deadlines under a Pareto contract) | smoke | citation-backed |
|
moe-expert-parallelmoe-expert-parallel a sparsely-activated mixture-of-experts model routes each token batch to one expert and the experts are spread across devices, and expert popularity is long-tailed, so a few hot experts receive most token batches while many stay cold and the all-to-all routing exchange dominates fabric traffic; the scheduler reasons over an expert-placement and load-balancing problem rather than a dense uniform-replica serving one | distributed-computing | ioe-complete | published dense generative-model serving systems (uniform per-replica KV-cache admission, not sparse token-to-expert routing under load imbalance and a Pareto contract) | smoke | citation-backed |
|
multi-objective-pfspmulti-objective-pfsp | flow-shop | classical | — | smoke | citation-backed |
|
multi-project-rcpspmulti-project-rcpsp | rcpsp | structurally-complex | — | smoke | citation-backed |
|
multi-tenant-fair-sharemulti-tenant-fair-share a shared cluster serves several tenants whose workloads compete for one node pool; each task is placement-flexible, carrying one single-node mode per pool node, so the scheduler chooses which node it occupies; tenants are sized asymmetrically, so even a load-balanced placement leaves the heavy tenants holding a larger fraction of their busiest node -- a higher dominant resource share -- than the light ones, and a fairness-aware scheduler rebalances placement to shrink the dominant-share spread | distributed-computing | ioe-complete | published multi-tenant colocation families (Borg-style priority colocation, vm allocation) that fix each task's resource and score makespan or cost; none lets the scheduler choose each tenant task's node and scores the dominant-resource-share spread between tenants as a fairness objective | smoke | citation-backed |
|
network-slicingnetwork-slicing isolated slice classes (latency-critical, broadband, massive-IoT) each carry their own service-level deadline and placement, and same-class slices spread across tiers for resilience | distributed-computing | ioe-complete | 5G slicing orchestration (no joint Pareto placement) | smoke | citation-backed |
|
no-wait-flow-shopno-wait-flow-shop | flow-shop | classical | — | smoke | citation-backed |
|
open-shopopen-shop | open-shop | classical | — | smoke | citation-backed |
|
orbital-edgeorbital-edge tasks schedule across ground terminals, moving low-earth-orbit satellites, and cloud backhaul under time-varying connectivity as satellites enter and leave coverage and hand work over | distributed-computing | ioe-complete | LENS real-measurement LEO satellite-network traces | smoke | citation-backed |
|
pipeline-parallel-trainingpipeline-parallel-training a deep network is split into successive pipeline stages pinned across edge-to-cloud tiers and the training mini-batch is divided into micro-batches, so each micro-batch flows forward stage by stage while each stage runs its micro-batches in issue order; the two precedence families form a diagonal wavefront whose warm-up and cool-down idle slots are the pipeline bubbles, and deeper stages carry rising compute, so the scheduler reasons over a stage-partition and bubble-minimizing problem rather than a synchronous data-parallel all-reduce one | distributed-computing | ioe-complete | published data-parallel / gang-scheduled training systems (synchronous all-reduce over co-located replicas, not a stage-by-micro-batch pipeline wavefront with warm-up and cool-down bubbles under a Pareto contract) | smoke | citation-backed |
|
rcpsprcpsp-renewable | rcpsp | structurally-complex | — | smoke | citation-backed |
|
rcpsp-maxrcpsp-max | rcpsp-max | structurally-complex | — | smoke | citation-backed |
|
rcpsp-multi-modercpsp-multi-mode | rcpsp | structurally-complex | — | smoke | citation-backed |
|
reentrant-fabreentrant-fab | job-shop | structurally-complex | — | smoke | citation-backed |
|
replica-placementreplica-placement a replica runs where its data shard already lives | distributed-computing | ioe-complete | CRUSH replicated-data placement | smoke | citation-backed |
|
serverless-cold-startserverless-cold-start a cold invocation pays a container provisioning penalty | distributed-computing | ioe-complete | Azure Functions serverless-in-the-wild traces | smoke | citation-backed |
|
service-function-chainservice-function-chain an NFV packet flow traverses a linear ordered chain of typed virtual network functions -- firewall, intrusion detection, deep packet inspection, address translation -- each pinned to a tier that hosts its function type, so the chain is a strict total order and the flow crosses the edge-fog-cloud fabric in a fixed sequence; the scheduler reasons over a chain-placement problem under an end-to-end latency budget rather than the branching role-pinned call graph of a microservice | distributed-computing | ioe-complete | published microservice call-graph schedulers (a branching role-pinned acyclic call graph, not a strict linear chain of function-typed network functions under an end-to-end latency budget and a Pareto contract) | smoke | citation-backed |
|
setup-flow-shopsetup-flow-shop | setup-flow-shop | classical | — | smoke | citation-backed |
|
smartnic-offloadsmartnic-offload a SmartNIC-accelerated server pairs a fast host CPU with a low-power on-NIC processor, and every microservice draws a long-tailed compute intensity, so most are light enough to offload onto the energy-frugal NIC cores while a few compute-heavy services must stay host-bound; a service's runtime scales with its intensity, so the scheduler reasons over an energy-versus-latency offload-placement problem rather than a uniform host placement one | distributed-computing | ioe-complete | published mobile-edge computation-offloading models (device-to-edge latency offload, not in-server host-to-NIC energy offload under a Pareto contract) | smoke | citation-backed |
|
split-inference-servingsplit-inference-serving each inference request partitions a deep model at a layer cut -- a light head runs the early layers on the edge near the sensor and a heavy tail runs the later layers in the cloud, consuming the head's intermediate feature map under a per-request end-to-end latency SLO | distributed-computing | ioe-complete | datacenter inference serving (no edge-cloud partition placement) | smoke | citation-backed |
|
spot-preemptiblespot-preemptible a cloud provider rents idle capacity at a discount as revocable spot instances reclaimed after a short lease; eviction-tolerant batch work pins to the cloud spot tier under a hard lease deadline (the eviction horizon), while latency-critical interactive work pins to the stable edge on-demand tier with no eviction deadline | distributed-computing | ioe-complete | cloud spot schedulers (no continuum eviction-deadline placement) | smoke | citation-backed |
|
storage-io-tieringstorage-io-tiering a tiered storage pool serves I/O-bound jobs whose runtime is dominated by moving a job's I/O volume through the storage node it lands on; each job is placement-flexible, carrying one single-node mode per tier node, and the mode duration is tier-dependent -- a seek floor plus the I/O volume divided by that tier's I/O bandwidth, which differs by tier (a fast cloud array sustains far more bytes/second than a slow edge disk); a job's I/O volume follows a heavy-tailed falloff over its I/O-demand rank, so a few I/O-heavy jobs carry most of the bytes and have a large cross-tier duration spread, while the light tail barely varies; a naive placement strands an I/O-heavy job on a low-bandwidth tier and pays its volume slowly, while a bandwidth-aware placement pins the heavy jobs to fast tiers to shrink makespan and cost | distributed-computing | ioe-complete | published storage-tiering and hierarchical-storage-management families that migrate blocks between fast and slow tiers by access frequency; none models the heterogeneous-tier I/O bandwidth as a placement-flexible per-tier mode whose duration is the I/O volume divided by that tier's bandwidth, so an I/O-heavy job's cross-tier duration spread is the bottleneck-relief lever a bandwidth-aware placement pulls under the Pareto contract | smoke | citation-backed |
|
streaming-windowstreaming-window events arrive online in bounded windows and must close within one | distributed-computing | ioe-complete | Parallel Workloads Archive online arrivals | smoke | citation-backed |
|
time-sensitive-networkingtime-sensitive-networking each time-triggered flow releases on a fixed period and must finish within one cycle under a hard, jitter-free deadline | distributed-computing | ioe-complete | EdgeCloudSim best-effort scenarios (no gating) | smoke | citation-backed |
|
unrelated-parallel-setupunrelated-parallel-setup | machine-scheduling | structurally-complex | — | smoke | citation-backed |
|
vehicular-offloadingvehicular-offloading a vehicle's tasks share an arrival time and a roadside-unit dwell deadline, and hand over from the roadside unit to the fog tier as the vehicle drives on | distributed-computing | ioe-complete | EdgeCloudSim / SUMO vehicular-edge mobility scenarios | smoke | citation-backed |
|
video-analyticsvideo-analytics each camera streams frames that must be analyzed within a tight real-time latency bound, placed hierarchically with edge inference near the camera and cloud aggregation | distributed-computing | ioe-complete | edge video-analytics clusters (no joint Pareto placement) | smoke | citation-backed |
|
vm-allocationvm-allocation size-heterogeneous virtual-machine deployments pack onto hosts while each deployment's members spread across distinct failure domains for availability | distributed-computing | ioe-complete | Azure Public Dataset Resource Central VM-allocation traces | smoke | citation-backed |
|
workflow-dagworkflow-dag | distributed-computing | structurally-complex | — | smoke | citation-backed |
|
Per-instance characterization and download eligibility live in the benchmark catalog; this table is the family-and-citation inventory.
El componente de arriba lleva el inventario generado completo — las familias núcleo clásicas y de distributed-computing de abajo más el continuo de familias Edge–Fog–Cloud. Esas familias núcleo fundacionales en forma de documento:
| Familia | Perfil | Clase de perfil | Corpus primario |
|---|---|---|---|
| Planificación de máquinas (R||Cmax, máquina no-relacionada) | machine-scheduling-unrelated | classical | OR-Library |
| Job-shop | job-shop-classical | classical | OR-Library, Taillard |
| Job-shop flexible (FJSP) | flexible-job-shop | classical | Brandimarte; Hurink-Jurisch-Thole |
| Flow-shop de permutación | permutation-flow-shop | classical | Taillard |
| Flow-shop de setup dependiente-de-secuencia (SDST) | setup-flow-shop | classical | Allahverdi et al. (2008); Allahverdi (2015) |
| Planificación de proyectos con-recursos-restringidos (RCPSP) | rcpsp-renewable | structurally-complex | PSPLIB |
| Distributed-computing (cloud/edge) | cloud-edge-capacity | domain-specific | CloudSim; DynamicCloudSim; Edge vision |
| Distributed-computing (workflow DAG) | workflow-dag | structurally-complex | Standard Task Graph Set |
Las familias de máquina-no-relacionada y job-shop flexible adjuntan una matriz de
tiempo-de-ejecución CostModel a cada instancia, exportada como una matriz materializada
junto al JSON del problema. La familia de flow-shop de setup dependiente-de-secuencia en
cambio adjunta una matriz de setup CostModel: un cambio entre trabajos de familias
diferentes en una máquina cuesta tiempo de setup, de modo que el objetivo de setup
recompensa agrupar trabajos similares. Los corpora estándar nombrados se citan y enlazan
solamente y nunca se redistribuyen dentro del repositorio.
Varias familias del continuo ejercitan co-asignación multi-recurso: cada tarea
demanda más de un recurso a la vez y el constructor los mantiene juntos durante toda su
duración (ver Contratos de dominio). La familia
accelerator-coscheduling co-asigna un nodo de cómputo y un acelerador escaso por
trabajo; la familia distributed-transaction co-asigna un conjunto de bloqueos de
cardinalidad-variable de fragmentos de datos por transacción; y la familia
fpga-partitioning co-asigna una corrida espacialmente contigua de tiles de tejido
reconfigurable por kernel de inquilino. Las tareas cuyos conjuntos de recursos se
intersectan se serializan mientras las tareas disjuntas corren concurrentemente — el
ejecutable examples/inspect_coallocation.py
hace la palanca explícita.
Más allá de la co-asignación, tres familias del continuo ejercitan sus propias palancas
estructurales. La familia elastic-serverless-autoscale ejercita ejecución moldeable:
cada invocación de función declara más de un modo de ejecución — un modo angosto
solo-en-casa y un modo ancho que toma prestado un worker de un pequeño pool de ráfaga
compartido para terminar antes — de modo que el horario elige un modo por tarea y el orden
decide qué invocaciones reclaman el escaso modo ancho-y-rápido. La familia
distributed-training-gang ejercita co-planificación de pandilla: los workers de un
trabajo de entrenamiento data-paralelo síncrono comparten una pandilla y deben co-iniciar
en aceleradores distintos en un lanzamiento todo-o-nada — los workers reutilizan el pool
de aceleradores y los trabajos llegan con el tiempo, de modo que un trabajo no puede
comenzar hasta que suficientes aceleradores se liberen simultáneamente, y el orden decide
qué trabajo adquiere su conjunto completo de workers primero. La familia
multi-tenant-fair-share ejercita equidad de recurso-dominante: varios inquilinos de
tamaño-asimétrico colocan tareas flexibles-en-colocación en un pool de nodos compartido, y
el objetivo de cuota-de-recurso-dominante puntúa la dispersión entre la cuota dominante
del inquilino más- y menos-servido — de modo que la colocación, qué recursos ocupa cada
inquilino, es la palanca que la equilibra o la sesga. Los ejecutables
examples/serverless_autoscale_study.py,
examples/distributed_training_gang_study.py,
y
examples/multi_tenant_fairshare_study.py
hacen estas tres palancas explícitas.
Clases de perfil
| Clase de perfil | Significado |
|---|---|
classical | Deriva de un corpus estándar de optimización-combinatoria. |
structurally-complex | Lleva estructura de precedencia, DAG, o red-de-recursos. |
ioe-complete | Escenario distribuido Internet-of-Everything-completo. |
trace-backed | Fundamentado en una traza de carga del mundo-real nombrada. |
domain-specific | Adaptado a un único dominio operacional. |
Etiquetas de evidencia
| Etiqueta | Uso |
|---|---|
smoke | Instancias deterministas pequeñas para pruebas, ejemplos, docs, y vistas previas. |
exploratory | Material plausible que aún no está respaldado-por-citaciones ni completamente caracterizado. |
| grado de evidencia candidato | Material respaldado-por-citaciones a la espera de gates de piloto, estadísticos, y de campaña. |
| grado de evidencia de campaña-completa | Evidencia que ha pasado los gates de citación, caracterización, estadísticos, de divulgación, y de calidad. |
Los catálogos de prueba nunca son evidencia de evaluación final. Existen para probar que
los generadores, la validación, la caracterización, las comprobaciones de citación, y la
persistencia funcionan rápidamente. El catálogo del portal y las descargas previsualizan
cada familia del continuo a esta pequeña escala de prueba (un pool de tres-recursos); una
familia de co-asignación no puede exhibir paralelismo de recurso-disjunto en un pool tan
pequeño, de modo que la estructura distintiva es una propiedad de escala-de-investigación.
build_continuum_full_catalog() materializa cada familia en su escala de investigación
declarada -- el pool de recursos y el conteo de tareas mayores donde la estructura de
co-asignación, contención, y colocación genuinamente se manifiesta -- para paquetes de
benchmarks de grado-investigación.
Taxonomía
La taxonomía cubre estructuras de planificación, entornos, realismo de infraestructura, características de objetivo, características de restricción, incertidumbre, y dinamismo. Los ejemplos incluyen workflows DAG, lotes de tareas independientes, funciones serverless, consolidación de contenedores y VM, entornos edge y cloud, trazas públicas, optimización multi-objetivo, deadlines, localidad de datos, churn, y arribos dinámicos.
Matriz de citación
Las afirmaciones de benchmarks se comprueban contra CitationMatrix. Las afirmaciones de
grado de evidencia candidato y de campaña-completa fallan la validación a menos que
referencien fuentes respaldadas-por-citaciones. El material no soportado debe permanecer
exploratorio hasta que se añada evidencia.
El conjunto de fuentes se declara en default_citation_matrix() con identificadores de
fuente estables, referencias resolubles, y una postura de licencia registrada. Abarca tres
niveles: los corpora estándar de optimización-combinatoria (citados y enlazados solamente,
nunca empaquetados), trazas de clúster de producción, y un amplio conjunto de datasets
contemporáneos del mundo-real del continuo Edge–Fog–Cloud — trazas de clúster de GPU y
machine-learning, suites de benchmark de microservicios y serverless, trazas de
workflow-científico, trazas de trabajos de supercomputador, datasets de edge-placement y
movilidad, datasets de IoT y demanda-celular, señales de carbono y energía de red, cargas
de procesamiento-de-flujo, benchmarks de participación-de-dispositivos de
federated-learning, testbeds de sistemas de realidad-extendida, y trazas de red-satelital
de órbita-terrestre-baja:
Suites de referencia canónicas
El registro en default_reference_suites() registra las suites de instancias publicadas
canónicas a las que ancla cada familia genérica de planificación: identidad, familia,
conteo de instancias, puntero de recuperación, y los trackers de
mejores-soluciones-conocidas que publican cotas para la suite. Las suites clásicas se
citan y enlazan solamente — DispatchAtlas nunca empaqueta ni redistribuye archivos de
instancias de terceros.
| Suite | Familia | Instancias | Descarga | Tracker de BKS |
|---|---|---|---|---|
fisher-thompson | job-shop | 3 | OR-Library | van-hoorn-2018, scheduleopt-benchmarks |
lawrence | job-shop | 40 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
adams-balas-zawack | job-shop | 5 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
applegate-cook-orb | job-shop | 10 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
storer-wu-vaccari | job-shop | 20 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
yamada-nakano | job-shop | 4 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
taillard-jsp | job-shop | 80 | Espejo JSPLIB | van-hoorn-2018, scheduleopt-benchmarks |
demirkol-dmu | job-shop | 80 | Espejo JSPLIB | scheduleopt-benchmarks |
brandimarte-mk | job-shop (flexible) | 15 | Espejo SchedulingLab | scheduleopt-benchmarks |
hurink-fjsp | job-shop (flexible) | 198 | Espejo SchedulingLab | scheduleopt-benchmarks |
dauzere-peres-paulli | job-shop (flexible) | 18 | Espejo SchedulingLab | scheduleopt-benchmarks |
taillard-pfsp | flow-shop | 120 | OR-Library | zenodo-pfsp-bks-2021 |
vrf-pfsp | flow-shop | 480 | Sitio del grupo SOA | zenodo-pfsp-bks-2021 |
sdst-taillard-ruiz | setup-flow-shop | 480 | Sitio del grupo SOA | las mejores soluciones acompañan a las instancias |
cicirello-wt-sds | machine-scheduling | 120 | Harvard Dataverse | cicirello-wtsds-benchmark |
or-library-smtwt | machine-scheduling | 375 | OR-Library | crauwels-potts-vanwassenhove-1998 |
vallada-ruiz-upmsp | machine-scheduling | 1640 (reportado) | Sitio del grupo SOA | — |
psplib | rcpsp | 2040 | Sitio de PSPLIB | psplib-1997 |
mmlib | rcpsp | 4320 (reportado) | Página de aterrizaje de OR&S | solutionsupdate-ugent-rcpsp |
rg300 | rcpsp | 480 | Página de aterrizaje de OR&S | solutionsupdate-ugent-rcpsp |
Las suites con un parser empaquetado (texto de job-shop estándar, matrices de flow-shop
de Taillard, job-shop flexible .fjs, JSON WfFormat de WfCommons) se ingieren con
load_reference_suite(suite_id, instances_root=...) desde archivos que el operador
descarga y coloca bajo un árbol resources/ local. La ingestión corre completamente
offline, reutiliza la misma validación, caracterización, hashing, y sobre de procedencia
que los generadores sintéticos, y estampa cada problema con su suite_id y su
upstream_instance_id. Las suites solo-de-registro se registran con sus citaciones y
punteros de recuperación sin un parser empaquetado.
Registros de mejores-soluciones-conocidas
Los valores mejores-conocidos por-instancia nunca se envían con DispatchAtlas. El
operador los ingiere como archivos JSON bajo un directorio privado
resources/benchmarks/bks/, un archivo por suite, cada uno llevando schema_version,
el suite_id, el source_id del tracker, la fecha de recuperación, y las entradas de
valor (identificador de instancia, objetivo, valor, tipo de óptimo-o-cota-superior, cota
inferior opcional). load_best_known_registry valida cada archivo contra las suites de
referencia y la matriz de citación y falla en cerrado ante suites desconocidas, trackers
desconocidos, entradas duplicadas, o cotas inconsistentes. Sin un registro ingerido, las
métricas de desviación-relativa simplemente no están disponibles — nunca se computan
parcialmente, y ningún valor mejor-conocido aparece en ninguna superficie pública.
Divergencias de calibración
Las familias genéricas sintéticas están ancladas a las suites canónicas sin pretender reproducir sus esquemas de generación. Las divergencias conocidas se documentan en lugar de ocultarse:
| Familia | Convención publicada | Convención sintética |
|---|---|---|
| flow-shop de setup | setups SDST-Taillard al 10/50/100/125% del tiempo de procesamiento | tres familias de setup, costo = familia + 1 |
| planificación de máquinas (R||Cmax) | clases de duración U[1,100] y variantes de máquina-correlacionada | factores de velocidad por-par 0.5–2.0 |
Las comparaciones contra las convenciones publicadas se enrutan a través de las instancias canónicas ingeridas, no a través de las familias sintéticas.
Caracterización
Cada problema materializado recibe descriptores normalizados para:
- densidad de oportunidad
- esparsidad de compatibilidad
- contención y sobrecarga
- profundidad de dependencia
- presión de comunicación
- intensidad de setup
- sesgo de carga y heterogeneidad
- conflicto de objetivo
- incertidumbre y dinamismo
- sensibilidad del solver
Puente de distribution-distance
Cada perfil de grado de evidencia de campaña-completa declara un puente de distribution-distance: su
estado (synthetic, calibrated-synthetic, trace-backed, o externally-sourced),
evidencia de calibración, escenario de dominio, cobertura de transferencia y disrupción, y
riesgo de distribution-distance residual. La promoción al grado de evidencia de campaña-completa
falla en cerrado a menos que se declaren las coberturas de transferencia y disrupción, y un
perfil calibrated-synthetic debe nombrar la referencia trace-backed contra la que calibra.
Las familias calibrated-synthetic nombran su referencia de traza explícitamente; las
suites canónicas ingeridas llevan un puente externally-sourced, y el adaptador de
WfCommons es la primera fuente de instancias trace-backed parseada-externamente, dando a
distribution_distance_score una pata de referencia trace-backed real.
La métrica de calibración nombrada reporta la distancia 1-Wasserstein (movedor-de-tierra)
por-característica entre las distribuciones de características de caracterización de un
perfil calibrated-synthetic y las de sus instancias de referencia trace-backed. Las
distancias por-característica se agregan a una única puntuación de distribution-distance; una
puntuación por encima del umbral de divergencia-máxima (por defecto 0.25) significa que
el perfil ha derivado demasiado lejos de su referencia y falla la calibración. La métrica
es reproducible desde las instancias materializadas y la traza de referencia nombrada.
from dispatchatlas.bench import (
build_smoke_catalog,
metrics_from_instance_set,
distribution_distance_score,
)
catalogs = {c.config.profile_id: c for c in build_smoke_catalog()}
synthetic = metrics_from_instance_set(catalogs["cloud-edge-capacity"])
reference = metrics_from_instance_set(catalogs["workflow-dag"])
score = distribution_distance_score(
synthetic, reference, reference_trace_id="google-cluster-data"
)Estratificación y selección de subconjunto
difficulty_score agrega los descriptores de contención, sobrecarga,
profundidad-de-dependencia, y sensibilidad-del-solver en una puntuación de dificultad
normalizada, y stratify_instances bina las instancias materializadas en estratos de
dificultad baja, media, y alta. select_benchmark_subset elige un subconjunto
determinista filtrado por familia, clase de perfil, y estrato de dificultad, ordenado por
identificador de problema de modo que la selección es reproducible.
Catálogo de prueba
from dispatchatlas.bench import build_smoke_catalog, smoke_benchmark_provider
catalogs = build_smoke_catalog(root_seed=20260527)
provider = smoke_benchmark_provider(root_seed=20260527)
first_problem = provider.get_problem(provider.list_problem_ids()[0])El catálogo de prueba empaquetado incluye las dos familias de distributed-computing (cloud/edge de tarea-independiente y workflow DAG) junto a las catorce familias genéricas de planificación (planificación de máquinas, job-shop, job-shop flexible, flow-shop de permutación, flow-shop de setup dependiente-de-secuencia, RCPSP, open-shop, flow-shop híbrido, flow-shop de permutación distribuido, flow-shop sin-espera, flow-shop con-bloqueo, flow-shop de ensamblaje distribuido, flow-shop de permutación multi-objetivo, y RCPSP/max) como pares co-iguales. Cada instancia es determinista desde la semilla raíz y usa metadatos de generador respaldados-por-citaciones mientras permanece etiquetada como material de desarrollo de prueba.
Catálogos completos candidatos
Los catálogos de campaña-completa candidatos usan la misma ruta de materialización con conteos de problema configurados mayores y etiquetas de evidencia más estrictas:
from dispatchatlas.bench import build_full_catalog, full_benchmark_provider
catalogs = build_full_catalog(root_seed=2026052713, problem_count_per_profile=30)
provider = full_benchmark_provider(
root_seed=2026052713,
problem_count_per_profile=30,
)Esos catálogos de grado de evidencia de campaña-completa están respaldados-por-citaciones, caracterizados, enlazados-por-hash, y aún marcados como evidencia de calibración hasta que los gates de campaña, divulgación, y publicación promuevan afirmaciones específicas.