{
  "artifact_hash": "f4f9f1c7128f60e96584fd594c1f7b7a2a1f3750fdccca4c90d669c121f99f26",
  "artifact_type": "dispatchatlas.benchmark_families",
  "evidence_class": "metadata-only",
  "families": [
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Ghodsi, Zaharia, Hindman, Konwinski, Shenker, and Stoica, USENIX NSDI 2011.",
          "source_id": "ghodsi-drf-2011",
          "uri": "https://www.usenix.org/conference/nsdi11/dominant-resource-fairness-fair-allocation-multiple-resource-types"
        },
        {
          "citation": "Xiao et al., USENIX OSDI 2018, 595-610.",
          "source_id": "gandiva-2018",
          "uri": "https://www.usenix.org/conference/osdi18/presentation/xiao"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "accelerator-coscheduling",
      "generator_id": "accelerator-coscheduling",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Sonmez, Ozgovde, and Ersoy, Transactions on Emerging Telecommunications Technologies, 29(11):e3493, 2018.",
          "source_id": "edgecloudsim-2018",
          "uri": "https://doi.org/10.1002/ett.3493"
        },
        {
          "citation": "Song, Qin, Hao, Hou, Wang, and Sun, arXiv:2208.13965, 2022.",
          "source_id": "aerial-mec-survey-2022",
          "uri": "https://arxiv.org/abs/2208.13965"
        }
      ],
      "closest_peer": "aerial-edge MEC simulators (no flying-fog loiter placement)",
      "evidence_class": "smoke",
      "family_id": "aerial-edge",
      "generator_id": "aerial-edge",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Liu, Lin, Shih, Yu, Chung, and Zhao, IEEE Computer 24(5):58-68, 1991.",
          "source_id": "imprecise-computation-1991",
          "uri": "https://doi.org/10.1109/2.76287"
        },
        {
          "citation": "Teerapittayanon, McDanel, and Kung, IEEE ICPR 2016, pp. 2464-2469.",
          "source_id": "branchynet-2016",
          "uri": "https://doi.org/10.1109/ICPR.2016.7900006"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "anytime-inference",
      "generator_id": "anytime-inference",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Hall and Sriskandarajah, Operations Research, 44(3):510-525, 1996.",
          "source_id": "hall-sriskandarajah-1996",
          "uri": "https://doi.org/10.1287/opre.44.3.510"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "blocking-flow-shop",
      "generator_id": "blocking-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Valiant, Communications of the ACM, 33(8):103-111, 1990.",
          "source_id": "valiant-bsp-1990",
          "uri": "https://doi.org/10.1145/79173.79181"
        },
        {
          "citation": "Malewicz, Austern, Bik, Dehnert, Horn, Leiser, and Czajkowski, ACM SIGMOD 2010, 135-146.",
          "source_id": "pregel-2010",
          "uri": "https://doi.org/10.1145/1807167.1807184"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "bulk-synchronous-graph",
      "generator_id": "bulk-synchronous-graph",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Electricity Maps, electricitymaps-contrib (open-source parsers and public grid carbon-intensity data).",
          "source_id": "electricity-maps-grid-ci",
          "uri": "https://github.com/electricitymaps/electricitymaps-contrib"
        },
        {
          "citation": "Nweye, Kaspar, Buscemi, et al., 2024 (arXiv:2405.03848); Vazquez-Canteli et al., ACM BuildSys 2019.",
          "source_id": "citylearn-v2",
          "uri": "https://github.com/intelligent-environments-lab/CityLearn"
        }
      ],
      "closest_peer": "Electricity-Maps grid carbon-intensity and CityLearn carbon-aware community signals",
      "evidence_class": "smoke",
      "family_id": "carbon-aware",
      "generator_id": "carbon-aware",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "flexible jobs defer to low-carbon-intensity windows under a time-varying grid carbon signal while honoring their SLA deadlines"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Calheiros, Ranjan, Beloglazov, De Rose, and Buyya, Software: Practice and Experience, 2011.",
          "source_id": "cloudsim-2011",
          "uri": "https://doi.org/10.1002/spe.995"
        },
        {
          "citation": "Bux and Leser, Future Generation Computer Systems, 2015.",
          "source_id": "dynamic-cloudsim-2015",
          "uri": "https://doi.org/10.1016/j.future.2014.09.007"
        },
        {
          "citation": "Shi, Cao, Zhang, Li, and Xu, IEEE Internet of Things Journal, 2016.",
          "source_id": "edge-vision-2016",
          "uri": "https://doi.org/10.1109/JIOT.2016.2579198"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "cloud-independent",
      "generator_id": "cloud-edge-independent",
      "profile_class": "domain-specific",
      "scheduling_family": "distributed-computing"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Chowdhury and Stoica, ACM Workshop on Hot Topics in Networks (HotNets) 2012, 31-36.",
          "source_id": "coflow-chowdhury-hotnets-2012",
          "uri": "https://doi.org/10.1145/2390231.2390237"
        },
        {
          "citation": "Chowdhury, Zhong, and Stoica, ACM SIGCOMM 2014, 443-454.",
          "source_id": "varys-chowdhury-sigcomm-2014",
          "uri": "https://doi.org/10.1145/2619239.2626315"
        }
      ],
      "closest_peer": "datacenter coflow schedulers (no continuum tier-placement barrier)",
      "evidence_class": "smoke",
      "family_id": "coflow-scheduling",
      "generator_id": "coflow-scheduling",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Beasley, Journal of the Operational Research Society, 41(11):1069-1072, 1990.",
          "source_id": "or-library-1990",
          "uri": "https://doi.org/10.1057/jors.1990.166"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "compact-job-shop",
      "generator_id": "compact-job-shop",
      "profile_class": "classical",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Shen, Tian, Chen, Chen, Wang, Xu, Yan, and Xia, International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2020.",
          "source_id": "occlum-asplos-2020",
          "uri": "https://arxiv.org/abs/2001.07450"
        },
        {
          "citation": "Zobaed and Amini Salehi, Software: Practice and Experience, 2025.",
          "source_id": "confidential-edge-zobaed-2025",
          "uri": "https://doi.org/10.1002/spe.3398"
        }
      ],
      "closest_peer": "edge enclave runtimes (no security-classified Pareto placement)",
      "evidence_class": "smoke",
      "family_id": "confidential-edge",
      "generator_id": "confidential-edge",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Liu and Layland, Journal of the ACM, 20(1):46-61, 1973.",
          "source_id": "liu-layland-1973",
          "uri": "https://doi.org/10.1145/321738.321743"
        },
        {
          "citation": "Logenthiran, Srinivasan, and Shun, IEEE Transactions on Smart Grid, 3(3):1244-1252, 2012.",
          "source_id": "smart-grid-dsm-2012",
          "uri": "https://doi.org/10.1109/TSG.2012.2195686"
        },
        {
          "citation": "Nweye, Kaspar, Buscemi, et al., 2024 (arXiv:2405.03848); Vazquez-Canteli et al., ACM BuildSys 2019.",
          "source_id": "citylearn-v2",
          "uri": "https://github.com/intelligent-environments-lab/CityLearn"
        }
      ],
      "closest_peer": "periodic hard-real-time task models and smart-grid demand-side scheduling formulations (no edge-fog-cloud tier placement under a Pareto contract)",
      "evidence_class": "smoke",
      "family_id": "cyber-physical",
      "generator_id": "cyber-physical",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "control cycles arrive periodically under a time-varying tariff"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Zaharia, Borthakur, Sen Sarma, Elmeleegy, Shenker, and Stoica, European Conference on Computer Systems (EuroSys) 2010, 265-278.",
          "source_id": "delay-scheduling-zaharia-eurosys-2010",
          "uri": "https://doi.org/10.1145/1755913.1755940"
        },
        {
          "citation": "Pu, Ananthanarayanan, Bodik, Kandula, Akella, Bahl, and Stoica, ACM SIGCOMM 2015, 421-434.",
          "source_id": "geo-analytics-iridium-pu-sigcomm-2015",
          "uri": "https://doi.org/10.1145/2829988.2787505"
        }
      ],
      "closest_peer": "cluster locality schedulers (no continuum data-residency placement)",
      "evidence_class": "smoke",
      "family_id": "data-locality",
      "generator_id": "data-locality",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Reiss, Wilkes, and Hellerstein, Google cluster-usage traces, 2011; Wilkes, ClusterData2019 v3, 2020.",
          "source_id": "google-cluster-data",
          "uri": "https://github.com/google/cluster-data"
        },
        {
          "citation": "Tirmazi, Barker, Deng, Haque, Qin, Hand, Harchol-Balter, and Wilkes, EuroSys 2020.",
          "source_id": "borg-2020",
          "uri": "https://doi.org/10.1145/3342195.3387517"
        }
      ],
      "closest_peer": "Google Borg ClusterData2019 priority-tiered cell traces",
      "evidence_class": "smoke",
      "family_id": "datacenter-colocation",
      "generator_id": "datacenter-colocation",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Tang, Chen, Koketsu Rodrigues, Zhao, and Kato, IEEE Open Journal of the Communications Society, 3:1360-1381, 2022.",
          "source_id": "digital-twin-diten-2022",
          "uri": "https://doi.org/10.1109/OJCOMS.2022.3197811"
        }
      ],
      "closest_peer": "digital-twin edge frameworks (no joint Pareto placement)",
      "evidence_class": "smoke",
      "family_id": "digital-twin-sync",
      "generator_id": "digital-twin-sync",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Li, Berger, Hsu, Ernst, Zardoshti, Novakovic, Shah, Rajadnya, Lee, Agarwal, Hill, Fontoura, and Bianchini, ASPLOS 2023.",
          "source_id": "pond-cxl-2023",
          "uri": "https://doi.org/10.1145/3575693.3578835"
        },
        {
          "citation": "Maruf, Wang, Dhanotia, Weiner, Agarwal, Bhattacharya, Petersen, Chowdhury, Kanaujia, and Chauhan, ASPLOS 2023.",
          "source_id": "tpp-cxl-2023",
          "uri": "https://doi.org/10.1145/3582016.3582063"
        }
      ],
      "closest_peer": "published CXL memory-pooling and tiered-memory systems (socket-local page placement, no edge-fog-cloud tier scheduling under a Pareto contract)",
      "evidence_class": "smoke",
      "family_id": "disaggregated-memory",
      "generator_id": "disaggregated-memory",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Hatami, Ruiz, and Andr\u00e9s-Romano, International Journal of Production Research, 51(17):5292-5308, 2013.",
          "source_id": "hatami-ruiz-2013",
          "uri": "https://doi.org/10.1080/00207543.2013.807955"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "distributed-assembly-flow-shop",
      "generator_id": "distributed-assembly-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "De Giovanni and Pezzella, European Journal of Operational Research, 200(2):395-408, 2010.",
          "source_id": "de-giovanni-pezzella-2010",
          "uri": "https://doi.org/10.1016/j.ejor.2009.01.008"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "distributed-flexible-job-shop",
      "generator_id": "distributed-flexible-job-shop",
      "profile_class": "structurally-complex",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Naderi and Ruiz, Computers & Operations Research, 37(4):754-768, 2010.",
          "source_id": "naderi-ruiz-2010",
          "uri": "https://doi.org/10.1016/j.cor.2009.06.019"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "distributed-permutation-flow-shop",
      "generator_id": "distributed-permutation-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Feitelson and Rudolph, Job Scheduling Strategies for Parallel Processing (JSSPP), LNCS 949, Springer, 1995, 1-18.",
          "source_id": "gang-scheduling-1995",
          "uri": "https://doi.org/10.1007/3-540-60153-8_20"
        },
        {
          "citation": "Goyal, Dollar, Girshick, Noordhuis, Wesolowski, Kyrola, Tulloch, Jia, and He, arXiv:1706.02677, 2017.",
          "source_id": "large-minibatch-sgd-2017",
          "uri": "https://arxiv.org/abs/1706.02677"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "distributed-training-gang",
      "generator_id": "distributed-training-gang",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Thomson, Diamond, Weng, Ren, Shao, and Abadi, ACM SIGMOD 2012.",
          "source_id": "calvin-2012",
          "uri": "https://doi.org/10.1145/2213836.2213838"
        },
        {
          "citation": "Corbett et al., USENIX OSDI 2012.",
          "source_id": "spanner-2012",
          "uri": "https://www.usenix.org/conference/osdi12/technical-sessions/presentation/corbett"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "distributed-transaction",
      "generator_id": "distributed-transaction",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Gupta, Vahid Dastjerdi, Ghosh, and Buyya, Software: Practice and Experience, 47(9):1275-1296, 2017.",
          "source_id": "ifogsim-2017",
          "uri": "https://doi.org/10.1002/spe.2509"
        },
        {
          "citation": "Mahmud, Pallewatta, Goudarzi, and Buyya, Journal of Systems and Software, 190:111351, 2022.",
          "source_id": "ifogsim2-2022",
          "uri": "https://doi.org/10.1016/j.jss.2022.111351"
        },
        {
          "citation": "Shi, Cao, Zhang, Li, and Xu, IEEE Internet of Things Journal, 2016.",
          "source_id": "edge-vision-2016",
          "uri": "https://doi.org/10.1109/JIOT.2016.2579198"
        },
        {
          "citation": "Wang, Guo, Zhang, Yang, Zhou, and Shen, IEEE Transactions on Mobile Computing, 20(3):939-953, 2021.",
          "source_id": "shanghai-telecom-edge",
          "uri": "https://wangshangguang.github.io/telecom_dataset/"
        }
      ],
      "closest_peer": "iFogSim MEC offloading scenarios",
      "evidence_class": "smoke",
      "family_id": "edge-offloading",
      "generator_id": "edge-offloading-mec",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "each task chooses between local edge execution and remote offload"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Lai, He, Abdelrazek, Chen, Bao, Grundy, Hosking, and Yang, ICSOC 2018.",
          "source_id": "eua-dataset",
          "uri": "https://github.com/swinedge/eua-dataset"
        },
        {
          "citation": "Wang, Guo, Zhang, Yang, Zhou, and Shen, IEEE Transactions on Mobile Computing, 20(3):939-953, 2021.",
          "source_id": "shanghai-telecom-edge",
          "uri": "https://wangshangguang.github.io/telecom_dataset/"
        }
      ],
      "closest_peer": "EUA edge-user-allocation and Shanghai-Telecom base-station placement traces",
      "evidence_class": "smoke",
      "family_id": "edge-placement",
      "generator_id": "edge-placement",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "services place on edge servers near their user population and migrate as demand shifts across base-station coverage cells"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Turek, Wolf, and Yu, ACM Symposium on Parallel Algorithms and Architectures (SPAA) 1992, 323-332.",
          "source_id": "moldable-parallelizable-tasks-1992",
          "uri": "https://doi.org/10.1145/140901.141909"
        },
        {
          "citation": "Castro, Ishakian, Muthusamy, and Slominski, Communications of the ACM 62(12):44-54, 2019.",
          "source_id": "serverless-rise-2019",
          "uri": "https://doi.org/10.1145/3368454"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "elastic-serverless-autoscale",
      "generator_id": "elastic-serverless-autoscale",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Hooker, Operations Research, 55(3):588-602, 2007.",
          "source_id": "hooker-2007",
          "uri": "https://doi.org/10.1287/opre.1060.0371"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "facility-assignment",
      "generator_id": "facility-assignment",
      "profile_class": "classical",
      "scheduling_family": "machine-scheduling"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Reiss, Wilkes, and Hellerstein, Google cluster-usage traces, 2011; Wilkes, ClusterData2019 v3, 2020.",
          "source_id": "google-cluster-data",
          "uri": "https://github.com/google/cluster-data"
        },
        {
          "citation": "Tirmazi, Barker, Deng, Haque, Qin, Hand, Harchol-Balter, and Wilkes, EuroSys 2020.",
          "source_id": "borg-2020",
          "uri": "https://doi.org/10.1145/3342195.3387517"
        }
      ],
      "closest_peer": "Borg cluster failure-event traces",
      "evidence_class": "smoke",
      "family_id": "failure-recovery",
      "generator_id": "failure-recovery",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "a failed task re-places its checkpoint state to a surviving tier"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Lai, Dai, Singapuram, Liu, Zhu, Madhyastha, and Chowdhury, ICML 2022 (PMLR 162); arXiv:2105.11367.",
          "source_id": "fedscale-2022",
          "uri": "https://github.com/SymbioticLab/FedScale"
        },
        {
          "citation": "Lai, Zhu, Madhyastha, and Chowdhury, USENIX OSDI 2021.",
          "source_id": "oort-2021",
          "uri": "https://github.com/SymbioticLab/Oort"
        }
      ],
      "closest_peer": "FedScale and Oort federated-learning device-participation benchmarks",
      "evidence_class": "smoke",
      "family_id": "federated-learning",
      "generator_id": "federated-learning",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Brandimarte, Annals of Operations Research, 41(3):157-183, 1993.",
          "source_id": "brandimarte-1993",
          "uri": "https://doi.org/10.1007/BF02023073"
        },
        {
          "citation": "Hurink, Jurisch, and Thole, OR Spektrum, 15(4):205-215, 1994.",
          "source_id": "hurink-jurisch-thole-1994",
          "uri": "https://doi.org/10.1007/BF01719451"
        },
        {
          "citation": "Berto, Hua, Park, et al., arXiv:2306.17100, 2023 (KDD 2025).",
          "source_id": "rl4co-2023",
          "uri": "https://arxiv.org/abs/2306.17100"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "flexible-job-shop",
      "generator_id": "flexible-job-shop",
      "profile_class": "classical",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        },
        {
          "citation": "Vallada, Ruiz, and Framinan, European Journal of Operational Research, 240(3):666-677, 2015.",
          "source_id": "vallada-ruiz-framinan-2015",
          "uri": "https://doi.org/10.1016/j.ejor.2014.07.033"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "flow-shop",
      "generator_id": "permutation-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Khawaja, Landgraf, Prakash, Wei, Schkufza, and Rossbach, USENIX OSDI 2018.",
          "source_id": "amorphos-2018",
          "uri": "https://www.usenix.org/conference/osdi18/presentation/khawaja"
        },
        {
          "citation": "Steiger, Walder, Platzner, and Thiele, IEEE RTSS 2003, pp. 224-235.",
          "source_id": "fpga-online-placement-2003",
          "uri": "https://doi.org/10.1109/REAL.2003.1253269"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "fpga-partitioning",
      "generator_id": "fpga-partitioning",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Feng, Sun, Li, Talwalkar, and Yang, arXiv:2505.16952, 2026 (ICLR 2026).",
          "source_id": "frontierco",
          "uri": "https://arxiv.org/abs/2505.16952"
        },
        {
          "citation": "Behnke and Geiger, Helmut-Schmidt-University Hamburg working paper, 2012.",
          "source_id": "behnke-geiger-2012",
          "uri": "https://doi.org/10.24405/436"
        },
        {
          "citation": "Naderi and Roshanaei, INFORMS Journal on Optimization, 4(1):1-28, 2022.",
          "source_id": "naderi-roshanaei-2022",
          "uri": "https://doi.org/10.1287/ijoo.2021.0056"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "frontierco-fjsp",
      "generator_id": "frontierco-fjsp",
      "profile_class": "classical",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Kwon, Li, Zhuang, Sheng, Zheng, Yu, Gonzalez, Zhang, and Stoica, SOSP 2023, pp. 611-626.",
          "source_id": "vllm-pagedattention-2023",
          "uri": "https://doi.org/10.1145/3600006.3613165"
        },
        {
          "citation": "Yu, Jeong, Kim, Kim, and Chun, OSDI 2022.",
          "source_id": "orca-serving-2022",
          "uri": "https://www.usenix.org/conference/osdi22/presentation/yu"
        }
      ],
      "closest_peer": "published single-replica generative-model serving systems (no edge-fog-cloud tier placement under a Pareto contract)",
      "evidence_class": "smoke",
      "family_id": "generative-inference-serving",
      "generator_id": "generative-inference-serving",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Weng, Xiao, Yu, et al., USENIX NSDI 2022.",
          "source_id": "alibaba-pai-gpu-2020",
          "uri": "https://github.com/alibaba/clusterdata/tree/master/cluster-trace-gpu-v2020"
        },
        {
          "citation": "Patel, Choukse, Zhang, Shah, Goiri, Maleki, and Bianchini, ISCA 2024.",
          "source_id": "azure-inference-trace-2023",
          "uri": "https://github.com/Azure/AzurePublicDataset/blob/master/AzureLLMInferenceDataset2023.md"
        },
        {
          "citation": "Jeon, Venkataraman, Phanishayee, Qian, Xiao, and Yang, USENIX ATC 2019.",
          "source_id": "philly-traces-2019",
          "uri": "https://github.com/msr-fiddle/philly-traces"
        },
        {
          "citation": "Hu, Sun, Yan, Wen, and Zhang, SC 2021.",
          "source_id": "helios-traces-2021",
          "uri": "https://github.com/S-Lab-System-Group/HeliosData"
        }
      ],
      "closest_peer": "Alibaba PAI, Philly, and Helios GPU-cluster traces",
      "evidence_class": "smoke",
      "family_id": "gpu-ml",
      "generator_id": "gpu-ml",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "training and inference jobs claim accelerators and gang-schedule replicas"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Ruiz and V\u00e1zquez-Rodr\u00edguez, European Journal of Operational Research, 205(1):1-18, 2010.",
          "source_id": "ruiz-vazquez-2010",
          "uri": "https://doi.org/10.1016/j.ejor.2009.09.024"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "hybrid-flow-shop",
      "generator_id": "hybrid-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Huzaifa, Desai, Grayson, et al., IEEE IISWC 2021.",
          "source_id": "illixr-2021",
          "uri": "https://github.com/ILLIXR/ILLIXR"
        }
      ],
      "closest_peer": "ILLIXR extended-reality systems testbed",
      "evidence_class": "smoke",
      "family_id": "immersive-xr",
      "generator_id": "immersive-xr",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Lucia and Ransford, ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI) 2015, 575-585.",
          "source_id": "batteryless-lucia-pldi-2015",
          "uri": "https://doi.org/10.1145/2737924.2737978"
        },
        {
          "citation": "Hester and Sorber, ACM Conference on Embedded Networked Sensor Systems (SenSys) 2017.",
          "source_id": "batteryless-hester-sensys-2017",
          "uri": "https://doi.org/10.1145/3131672.3131699"
        }
      ],
      "closest_peer": "intermittent-computing runtimes (no continuum energy-window placement)",
      "evidence_class": "smoke",
      "family_id": "intermittent-edge",
      "generator_id": "intermittent-edge",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Bodik, Hong, Guestrin, Madden, Paskin, and Thibaux, Intel Berkeley Research Lab / MIT CSAIL, 2004.",
          "source_id": "intel-lab-data",
          "uri": "https://db.csail.mit.edu/labdata/labdata.html"
        },
        {
          "citation": "Barlacchi, De Nadai, Larcher, et al., Scientific Data, 2:150055, 2015.",
          "source_id": "milano-cdr-2015",
          "uri": "https://doi.org/10.1038/sdata.2015.55"
        }
      ],
      "closest_peer": "published wireless-sensor-network telemetry datasets and in-network aggregation deployments (raw sensor readings, not edge-fog-cloud tier scheduling under a Pareto contract)",
      "evidence_class": "smoke",
      "family_id": "iot-edge",
      "generator_id": "iot-edge",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "many small sensor readings arrive periodically and aggregate at the edge"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Beasley, Journal of the Operational Research Society, 41(11):1069-1072, 1990.",
          "source_id": "or-library-1990",
          "uri": "https://doi.org/10.1057/jors.1990.166"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        },
        {
          "citation": "Applegate and Cook, ORSA Journal on Computing, 3(2):149-156, 1991.",
          "source_id": "applegate-cook-1991",
          "uri": "https://doi.org/10.1287/ijoc.3.2.149"
        },
        {
          "citation": "Demirkol, Mehta, and Uzsoy, European Journal of Operational Research, 109(1):137-141, 1998.",
          "source_id": "demirkol-mehta-uzsoy-1998",
          "uri": "https://doi.org/10.1016/S0377-2217(97)00019-2"
        },
        {
          "citation": "Storer, Wu, and Vaccari, Management Science, 38(10):1495-1509, 1992.",
          "source_id": "storer-wu-vaccari-1992",
          "uri": "https://doi.org/10.1287/mnsc.38.10.1495"
        },
        {
          "citation": "Berto, Hua, Park, et al., arXiv:2306.17100, 2023 (KDD 2025).",
          "source_id": "rl4co-2023",
          "uri": "https://arxiv.org/abs/2306.17100"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "job-shop",
      "generator_id": "job-shop",
      "profile_class": "classical",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Yang, Yue, and Rashmi, USENIX OSDI 2020; 54 in-memory cache clusters over one week.",
          "source_id": "twitter-cache-traces-2020",
          "uri": "https://www.usenix.org/conference/osdi20/presentation/yang"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "kv-cache-placement",
      "generator_id": "kv-cache-placement",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Beasley, Journal of the Operational Research Society, 41(11):1069-1072, 1990.",
          "source_id": "or-library-1990",
          "uri": "https://doi.org/10.1057/jors.1990.166"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "machine-scheduling",
      "generator_id": "machine-scheduling-unrelated",
      "profile_class": "classical",
      "scheduling_family": "machine-scheduling"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Alibaba cluster trace repository (v2017 / v2018-DAG / v2021-microservices).",
          "source_id": "alibaba-cluster-data",
          "uri": "https://github.com/alibaba/clusterdata"
        },
        {
          "citation": "Luo, Xu, Lu, Ye, et al., ACM SoCC 2021.",
          "source_id": "alibaba-microservice-characterization-2021",
          "uri": "https://doi.org/10.1145/3472883.3487003"
        },
        {
          "citation": "Zhou, Peng, Xie, et al., IEEE Transactions on Software Engineering, 2021.",
          "source_id": "train-ticket-benchmark",
          "uri": "https://github.com/FudanSELab/train-ticket"
        }
      ],
      "closest_peer": "Alibaba v2021 microservice-trace call graphs",
      "evidence_class": "smoke",
      "family_id": "microservice-dag",
      "generator_id": "microservice-dag",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "services form an acyclic call graph pinned by role to a tier"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Vestal, IEEE Real-Time Systems Symposium (RTSS) 2007, 239-243.",
          "source_id": "vestal-2007",
          "uri": "https://doi.org/10.1109/RTSS.2007.47"
        },
        {
          "citation": "Burns and Davis, ACM Computing Surveys, 50(6):82:1-82:37, 2018 (online 2017).",
          "source_id": "mixed-criticality-survey-2017",
          "uri": "https://doi.org/10.1145/3131347"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "mixed-criticality",
      "generator_id": "mixed-criticality",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Fedus, Zoph, and Shazeer, Journal of Machine Learning Research, 23(120):1-39, 2022.",
          "source_id": "switch-transformers-2022",
          "uri": "https://arxiv.org/abs/2101.03961"
        },
        {
          "citation": "Li et al., USENIX ATC 2023.",
          "source_id": "lina-moe-2023",
          "uri": "https://www.usenix.org/conference/atc23/presentation/li-jiamin"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "moe-expert-parallel",
      "generator_id": "moe-expert-parallel",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Minella, Ruiz, and Ciavotta, INFORMS Journal on Computing, 20(3):451-471, 2008.",
          "source_id": "minella-ruiz-2008",
          "uri": "https://doi.org/10.1287/ijoc.1070.0258"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "multi-objective-pfsp",
      "generator_id": "multi-objective-pfsp",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Van Eynde and Vanhoucke, Journal of Scheduling, 23(3):301-325, 2020.",
          "source_id": "van-eynde-vanhoucke-2020",
          "uri": "https://doi.org/10.1007/s10951-020-00651-w"
        },
        {
          "citation": "Kolisch and Sprecher, European Journal of Operational Research, 96(1):205-216, 1997.",
          "source_id": "psplib-1997",
          "uri": "https://doi.org/10.1016/S0377-2217(96)00170-1"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "multi-project-rcpsp",
      "generator_id": "multi-project-rcpsp",
      "profile_class": "structurally-complex",
      "scheduling_family": "rcpsp"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Ghodsi, Zaharia, Hindman, Konwinski, Shenker, and Stoica, USENIX NSDI 2011.",
          "source_id": "ghodsi-drf-2011",
          "uri": "https://www.usenix.org/conference/nsdi11/dominant-resource-fairness-fair-allocation-multiple-resource-types"
        },
        {
          "citation": "Tirmazi, Barker, Deng, Haque, Qin, Hand, Harchol-Balter, and Wilkes, EuroSys 2020.",
          "source_id": "borg-2020",
          "uri": "https://doi.org/10.1145/3342195.3387517"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "multi-tenant-fair-share",
      "generator_id": "multi-tenant-fair-share",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Afolabi, Taleb, Samdanis, Ksentini, and Flinck, IEEE Communications Surveys & Tutorials, 20(3):2429-2453, 2018.",
          "source_id": "network-slicing-afolabi-2018",
          "uri": "https://doi.org/10.1109/COMST.2018.2815638"
        }
      ],
      "closest_peer": "5G slicing orchestration (no joint Pareto placement)",
      "evidence_class": "smoke",
      "family_id": "network-slicing",
      "generator_id": "network-slicing",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Hall and Sriskandarajah, Operations Research, 44(3):510-525, 1996.",
          "source_id": "hall-sriskandarajah-1996",
          "uri": "https://doi.org/10.1287/opre.44.3.510"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "no-wait-flow-shop",
      "generator_id": "no-wait-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Gonzalez and Sahni, Journal of the ACM, 23(4):665-679, 1976.",
          "source_id": "gonzalez-sahni-1976",
          "uri": "https://doi.org/10.1145/321978.321985"
        },
        {
          "citation": "Taillard, European Journal of Operational Research, 1993.",
          "source_id": "taillard-1993",
          "uri": "https://doi.org/10.1016/0377-2217(93)90182-M"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "open-shop",
      "generator_id": "open-shop",
      "profile_class": "classical",
      "scheduling_family": "open-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Zhao and Pan, ACM MMSys 2024.",
          "source_id": "lens-2024",
          "uri": "https://github.com/clarkzjw/LENS"
        }
      ],
      "closest_peer": "LENS real-measurement LEO satellite-network traces",
      "evidence_class": "smoke",
      "family_id": "orbital-edge",
      "generator_id": "orbital-edge",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Huang, Cheng, Bapna, Firat, Chen, Chen, Lee, Ngiam, Le, Wu, and Chen, NeurIPS 2019.",
          "source_id": "gpipe-2019",
          "uri": "https://arxiv.org/abs/1811.06965"
        },
        {
          "citation": "Narayanan, Harlap, Phanishayee, Seshadri, Devanur, Ganger, Gibbons, and Zaharia, SOSP 2019.",
          "source_id": "pipedream-2019",
          "uri": "https://doi.org/10.1145/3341301.3359646"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "pipeline-parallel-training",
      "generator_id": "pipeline-parallel-training",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Kolisch and Sprecher, European Journal of Operational Research, 96(1):205-216, 1997.",
          "source_id": "psplib-1997",
          "uri": "https://doi.org/10.1016/S0377-2217(96)00170-1"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "rcpsp",
      "generator_id": "rcpsp-renewable",
      "profile_class": "structurally-complex",
      "scheduling_family": "rcpsp"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Bartusch, M\u00f6hring, and Radermacher, Annals of Operations Research, 16(1):199-240, 1988.",
          "source_id": "bartusch-moehring-1988",
          "uri": "https://doi.org/10.1007/BF02283745"
        },
        {
          "citation": "Kolisch and Sprecher, European Journal of Operational Research, 96(1):205-216, 1997.",
          "source_id": "psplib-1997",
          "uri": "https://doi.org/10.1016/S0377-2217(96)00170-1"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "rcpsp-max",
      "generator_id": "rcpsp-max",
      "profile_class": "structurally-complex",
      "scheduling_family": "rcpsp-max"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Van Peteghem and Vanhoucke, European Journal of Operational Research, 235(1):62-72, 2014.",
          "source_id": "van-peteghem-vanhoucke-2014",
          "uri": "https://doi.org/10.1016/j.ejor.2013.10.012"
        },
        {
          "citation": "Kolisch and Sprecher, European Journal of Operational Research, 96(1):205-216, 1997.",
          "source_id": "psplib-1997",
          "uri": "https://doi.org/10.1016/S0377-2217(96)00170-1"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "rcpsp-multi-mode",
      "generator_id": "rcpsp-multi-mode",
      "profile_class": "structurally-complex",
      "scheduling_family": "rcpsp"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Kopp, Hassoun, Kalir, and M\u00f6nch, IEEE Transactions on Semiconductor Manufacturing, 33(4):522-531, 2020.",
          "source_id": "smt2020-2020",
          "uri": "https://doi.org/10.1109/TSM.2020.3001933"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "reentrant-fab",
      "generator_id": "reentrant-fab",
      "profile_class": "structurally-complex",
      "scheduling_family": "job-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Weil, Brandt, Miller, and Maltzahn, Proceedings of the 2006 ACM/IEEE Conference on Supercomputing (SC'06), 2006.",
          "source_id": "crush-2006",
          "uri": "https://doi.org/10.1145/1188455.1188582"
        },
        {
          "citation": "Reiss, Wilkes, and Hellerstein, Google cluster-usage traces, 2011; Wilkes, ClusterData2019 v3, 2020.",
          "source_id": "google-cluster-data",
          "uri": "https://github.com/google/cluster-data"
        }
      ],
      "closest_peer": "CRUSH replicated-data placement",
      "evidence_class": "smoke",
      "family_id": "replica-placement",
      "generator_id": "replica-placement",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "a replica runs where its data shard already lives"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Cortez et al., SOSP'17 Resource Central; Shahrad et al., USENIX ATC'20 Serverless in the Wild; Hadary et al., USENIX OSDI'20 Protean (AzureTracesForPacking2020).",
          "source_id": "azure-public-dataset",
          "uri": "https://github.com/Azure/AzurePublicDataset"
        },
        {
          "citation": "Copik, Kwasniewski, Besta, Podstawski, and Hoefler, ACM/IFIP Middleware 2021.",
          "source_id": "sebs-2021",
          "uri": "https://doi.org/10.1145/3464298.3476133"
        }
      ],
      "closest_peer": "Azure Functions serverless-in-the-wild traces",
      "evidence_class": "smoke",
      "family_id": "serverless-cold-start",
      "generator_id": "serverless-cold-start",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "a cold invocation pays a container provisioning penalty"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Mijumbi, Serrat, Gorricho, Bouten, De Turck, and Boutaba, IEEE Communications Surveys & Tutorials, 18(1):236-262, 2016.",
          "source_id": "nfv-survey-2016",
          "uri": "https://doi.org/10.1109/COMST.2015.2477041"
        },
        {
          "citation": "Mehraghdam, Keller, and Karl, IEEE CloudNet 2014, 7-13.",
          "source_id": "sfc-placement-2014",
          "uri": "https://doi.org/10.1109/CloudNet.2014.6968961"
        }
      ],
      "closest_peer": "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)",
      "evidence_class": "smoke",
      "family_id": "service-function-chain",
      "generator_id": "service-function-chain",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Allahverdi, Ng, Cheng, and Kovalyov, European Journal of Operational Research, 187(3):985-1032, 2008.",
          "source_id": "allahverdi-survey-2008",
          "uri": "https://doi.org/10.1016/j.ejor.2006.06.060"
        },
        {
          "citation": "Allahverdi, European Journal of Operational Research, 246(2):345-378, 2015.",
          "source_id": "allahverdi-survey-2015",
          "uri": "https://doi.org/10.1016/j.ejor.2015.04.004"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "setup-flow-shop",
      "generator_id": "setup-flow-shop",
      "profile_class": "classical",
      "scheduling_family": "setup-flow-shop"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Liu, Cui, Schuh, Krishnamurthy, Peter, and Gupta, SIGCOMM 2019.",
          "source_id": "ipipe-smartnic-2019",
          "uri": "https://doi.org/10.1145/3341302.3342079"
        },
        {
          "citation": "Liu, Peter, Krishnamurthy, and Phothilimthana, USENIX ATC 2019.",
          "source_id": "e3-smartnic-2019",
          "uri": "https://www.usenix.org/conference/atc19/presentation/liu-ming"
        }
      ],
      "closest_peer": "published mobile-edge computation-offloading models (device-to-edge latency offload, not in-server host-to-NIC energy offload under a Pareto contract)",
      "evidence_class": "smoke",
      "family_id": "smartnic-offload",
      "generator_id": "smartnic-offload",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Li, Zheng, Zhong, Liu, Sheng, Jin, Huang, Chen, Zhang, Gonzalez, and Stoica, USENIX Symposium on Operating Systems Design and Implementation (OSDI) 2023, 663-679.",
          "source_id": "alpaserve-osdi-2023",
          "uri": "https://www.usenix.org/conference/osdi23/presentation/li-zhouhan"
        },
        {
          "citation": "Kang, Hauswald, Gao, Rovinski, Mudge, Mars, and Tang, International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2017.",
          "source_id": "neurosurgeon-asplos-2017",
          "uri": "https://doi.org/10.1145/3093337.3037698"
        }
      ],
      "closest_peer": "datacenter inference serving (no edge-cloud partition placement)",
      "evidence_class": "smoke",
      "family_id": "split-inference-serving",
      "generator_id": "split-inference-serving",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Ambati, Goiri, Frujeri, Gun, Wang, Dolan, Corell, Pasupuleti, Moscibroda, Elnikety, Fontoura, and Bianchini, USENIX Symposium on Operating Systems Design and Implementation (OSDI) 2020, 735-751.",
          "source_id": "harvest-vms-ambati-osdi-2020",
          "uri": "https://www.usenix.org/conference/osdi20/presentation/ambati"
        },
        {
          "citation": "Yang, Pang, Zhang, Qiao, Wang, Couturier, Bansal, Ram, Qin, Ma, Goiri, Cortez, Baladhandayutham, Ruhle, Rajmohan, Lin, and Zhang, The Web Conference (WWW) 2022 Companion, 152-156.",
          "source_id": "spot-eviction-yang-www-2022",
          "uri": "https://doi.org/10.1145/3487553.3524229"
        }
      ],
      "closest_peer": "cloud spot schedulers (no continuum eviction-deadline placement)",
      "evidence_class": "smoke",
      "family_id": "spot-preemptible",
      "generator_id": "spot-preemptible",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Narayanan, Donnelly, and Rowstron, ACM Transactions on Storage 4(3):10, 2008 (USENIX FAST 2008); MSR Cambridge enterprise volume block-I/O trace via the SNIA IOTTA repository.",
          "source_id": "msr-cambridge-block-2008",
          "uri": "https://doi.org/10.1145/1416944.1416949"
        }
      ],
      "closest_peer": "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",
      "evidence_class": "smoke",
      "family_id": "storage-io-tiering",
      "generator_id": "storage-io-tiering",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Feitelson, Tsafrir, and Krakov, Journal of Parallel and Distributed Computing, 74(10):2967-2982, 2014.",
          "source_id": "parallel-workloads-archive",
          "uri": "https://doi.org/10.1016/j.jpdc.2014.06.013"
        },
        {
          "citation": "Bordin, Griebler, Mencagli, Geyer, and Fernandes, IEEE Access, 8:222900-222917, 2020.",
          "source_id": "dspbench-2020",
          "uri": "https://doi.org/10.1109/ACCESS.2020.3043948"
        }
      ],
      "closest_peer": "Parallel Workloads Archive online arrivals",
      "evidence_class": "smoke",
      "family_id": "streaming-window",
      "generator_id": "streaming-window",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "events arrive online in bounded windows and must close within one"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Craciunas, Serna Oliver, Chmelik, and Steiner, RTNS 2016, ACM, 183-192.",
          "source_id": "tsn-craciunas-2016",
          "uri": "https://doi.org/10.1145/2997465.2997470"
        },
        {
          "citation": "Sonmez, Ozgovde, and Ersoy, Transactions on Emerging Telecommunications Technologies, 29(11):e3493, 2018.",
          "source_id": "edgecloudsim-2018",
          "uri": "https://doi.org/10.1002/ett.3493"
        }
      ],
      "closest_peer": "EdgeCloudSim best-effort scenarios (no gating)",
      "evidence_class": "smoke",
      "family_id": "time-sensitive-networking",
      "generator_id": "time-sensitive-networking",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "each time-triggered flow releases on a fixed period and must finish within one cycle under a hard, jitter-free deadline"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Vallada and Ruiz, European Journal of Operational Research, 211(3):612-622, 2011.",
          "source_id": "vallada-ruiz-2011",
          "uri": "https://doi.org/10.1016/j.ejor.2011.01.011"
        },
        {
          "citation": "Allahverdi, European Journal of Operational Research, 246(2):345-378, 2015.",
          "source_id": "allahverdi-survey-2015",
          "uri": "https://doi.org/10.1016/j.ejor.2015.04.004"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "unrelated-parallel-setup",
      "generator_id": "unrelated-parallel-setup",
      "profile_class": "structurally-complex",
      "scheduling_family": "machine-scheduling"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Codec\u00e0, Frank, Faye, and Engel, IEEE Intelligent Transportation Systems Magazine, 9(2):52-63, 2017.",
          "source_id": "lust-scenario",
          "uri": "https://github.com/lcodeca/LuSTScenario"
        },
        {
          "citation": "Sonmez, Ozgovde, and Ersoy, Transactions on Emerging Telecommunications Technologies, 29(11):e3493, 2018.",
          "source_id": "edgecloudsim-2018",
          "uri": "https://doi.org/10.1002/ett.3493"
        }
      ],
      "closest_peer": "EdgeCloudSim / SUMO vehicular-edge mobility scenarios",
      "evidence_class": "smoke",
      "family_id": "vehicular-offloading",
      "generator_id": "vehicular-offloading",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Hung, Ananthanarayanan, Bodik, Golubchik, Yu, Bahl, and Philipose, ACM/IEEE Symposium on Edge Computing (SEC) 2018, 115-131.",
          "source_id": "videoedge-2018",
          "uri": "https://doi.org/10.1109/SEC.2018.00016"
        }
      ],
      "closest_peer": "edge video-analytics clusters (no joint Pareto placement)",
      "evidence_class": "smoke",
      "family_id": "video-analytics",
      "generator_id": "video-analytics",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "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"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Cortez et al., SOSP'17 Resource Central; Shahrad et al., USENIX ATC'20 Serverless in the Wild; Hadary et al., USENIX OSDI'20 Protean (AzureTracesForPacking2020).",
          "source_id": "azure-public-dataset",
          "uri": "https://github.com/Azure/AzurePublicDataset"
        },
        {
          "citation": "Shen, van Beek, and Iosup, IEEE/ACM CCGrid 2015, pp. 465-474; Bitbrains (now Solvinity) datacenter VM trace via the Grid Workloads Archive (GWA-T-12).",
          "source_id": "bitbrains-gwa-t12-2015",
          "uri": "https://doi.org/10.1109/CCGRID.2015.60"
        },
        {
          "citation": "Kohne, Spohr, Nagel, and Spinczyk, ACM CCB@Middleware 2014; Materna GmbH datacenter VM trace via the Grid Workloads Archive (GWA-T-13).",
          "source_id": "materna-gwa-t13-2014",
          "uri": "https://doi.org/10.1145/2676662.2676674"
        },
        {
          "citation": "Beloglazov and Buyya, Concurrency and Computation: Practice and Experience 24(13):1397-1420, 2012; PlanetLab/CoMon CPU-utilization workload bundled with CloudSim.",
          "source_id": "planetlab-workload-2012",
          "uri": "https://doi.org/10.1002/cpe.1867"
        }
      ],
      "closest_peer": "Azure Public Dataset Resource Central VM-allocation traces",
      "evidence_class": "smoke",
      "family_id": "vm-allocation",
      "generator_id": "vm-allocation",
      "profile_class": "ioe-complete",
      "scheduling_family": "distributed-computing",
      "summary": "size-heterogeneous virtual-machine deployments pack onto hosts while each deployment's members spread across distinct failure domains for availability"
    },
    {
      "citation_status": "citation-backed",
      "citations": [
        {
          "citation": "Tobita and Kasahara, Journal of Scheduling, 5(5):379-394, 2002.",
          "source_id": "stg-task-graphs",
          "uri": "https://doi.org/10.1002/jos.116"
        },
        {
          "citation": "Coleman, Casanova, Pottier, Kaushik, Deelman, and Ferreira da Silva, Future Generation Computer Systems, 128:16-27, 2022.",
          "source_id": "wfcommons-pegasus-instances",
          "uri": "https://doi.org/10.5281/zenodo.4718070"
        },
        {
          "citation": "Topcuoglu, Hariri, and Wu, IEEE Transactions on Parallel and Distributed Systems, 13(3):260-274, 2002.",
          "source_id": "topcuoglu-heft-2002",
          "uri": "https://doi.org/10.1109/71.993206"
        },
        {
          "citation": "Canon, El Sayah, and H\u00e9am, Euro-Par 2019, LNCS 11725.",
          "source_id": "canon-dag-bias-2019",
          "uri": "https://doi.org/10.1007/978-3-030-29400-7_5"
        }
      ],
      "evidence_class": "smoke",
      "family_id": "workflow-dag",
      "generator_id": "workflow-dag",
      "profile_class": "structurally-complex",
      "scheduling_family": "distributed-computing"
    }
  ],
  "family_count": 69,
  "producer": "dispatchatlas.bench",
  "scheduling_families": [
    "distributed-computing",
    "flow-shop",
    "job-shop",
    "machine-scheduling",
    "open-shop",
    "rcpsp",
    "rcpsp-max",
    "setup-flow-shop"
  ],
  "scheduling_family_counts": {
    "distributed-computing": 47,
    "flow-shop": 7,
    "job-shop": 6,
    "machine-scheduling": 3,
    "open-shop": 1,
    "rcpsp": 3,
    "rcpsp-max": 1,
    "setup-flow-shop": 1
  },
  "schema_version": "1.0"
}