logseq/pages/hls__ECMFA_23_paper_8566_1680594411934_0.md

file:: ECMFA_23_paper_8566_1680594411934_0.pdf file-path:: ../assets/ECMFA_23_paper_8566_1680594411934_0.pdf

• design and management ls-type:: annotation hl-page:: 1 hl-color:: green id:: 642bd6f8-e260-4491-8d11-89d70de0c423
• ollect real-time data and control operation ls-type:: annotation hl-page:: 1 hl-color:: green id:: 642bd774-35c5-4be6-84cd-d4e2984011c8
• Additionally, 20 multi-layer IoT systems that leverage edge and fog computing 21 also deploy nodes or compute units (located in the plant) to run 22 lightweight applications and cloud servers for the deployment 23 and execution of the resource-intensive applications ls-type:: annotation hl-page:: 1 hl-color:: green id:: 642bed44-a177-49f5-b60f-dabeb014185e
• challenges related to the3 design of these block diagrams and the IoT system involved. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf621-0c74-4d31-861f-4ad7e4c59045
• unified language for modeling WWTP5 process block diagram ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf627-a769-4c94-bdc8-5b8291af5e80
• management and adaptation of the system to9 address functional requirements and quality of service (QoS) at10 run-time. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf631-6af5-4272-97a8-e1336c10306a
• process block diagram of a WWTP ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf643-f695-4883-a901-63bb28ecf994
• A code generator and MAPE-K based framework. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf651-e5c8-48b3-b477-ec7bfce1f11d
• A 81 model-based infrastructure for the specification and runtime 82 execution of self-adaptive IoT architectures. ls-type:: annotation hl-page:: 12 hl-color:: green id:: 642bf65f-b066-4b87-b337-953e0e11dbde
• design 43 time for the specification of the WWTP process block diagram ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf686-de7d-4b6a-8657-fe636e55c453
• self-adaptive IoT system, and run-time to support the 45 operation and adaptation of the system. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf68d-227b-41ab-a468-38ab3a523364
• multi-layer 54 IoT architectures ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf6b8-8f6e-4d86-82bc-02110a307f1f
• WWTP processes and the IoT system 59 implied ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf718-3f6e-407d-b6ed-9573210b14ff
• Both the DSL and the code 64 generator are implemented using MPS 1, a language workbench 65 developed by JetBrains to design DSLs. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 642bf725-6800-4886-b57d-9522bef5ea46
• Prometheus2—a time-series database ls-type:: annotation hl-page:: 3 hl-color:: green id:: 642bf743-e011-4b68-929e-4b6aab626c67
• WTP process block diagram — We enable the specifica-26 tion of the main operating units such as filters, grit cham-27 bers, biological reactors, and the flow—either water or28 sludge—between these operating units. ls-type:: annotation hl-page:: 3 hl-color:: green id:: 642bf77a-2c7c-4ae0-afc2-7b6b39e24e14
• first poin ls-type:: annotation hl-page:: 3 hl-color:: green id:: 642bf77f-e388-4f93-b194-92bdebf5e52a
• adap- 37 tation rules ls-type:: annotation hl-page:: 3 hl-color:: yellow id:: 642bf791-b109-4ad3-819e-37a29c10523f
• igure 3 defines the concepts for16 the specification of functional and adaptation rules ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c1a5a-c96f-47f1-af35-ed46c2cb892a
• A Rule is composed of an Expression—condition relation-18 ship—and multiple Actions that are executed on the system if19 the condition is true during a defined perio ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c1a67-5553-4551-a434-6aeede1d9310
• Expression concept by adding sensor and QoS25 conditions that can be combined also with all other types of26 expressions—e.g., BinaryOperations, Literals, or BooleanCon-27 stants—in a complex conditional expression. ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c1a96-6b20-44e8-a5c3-36b0fc2bc38b
• QoSCondition allows detecting unusual values 40 in QoS and system infrastructure metrics. For example, high 41 RAM and CPU consumption in one specific node (edge, fog, or 42 cloud), or in a group of nodes belonging to a Region. ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23ab-58bb-4cda-a136-1c411a0d385a
• Redeployment ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23b0-632f-4f0a-9ff3-4af1d087d4ee
• Offloading ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23c0-6e75-40b2-9fd3-e97757e68f15
• Scaling ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23c4-7e8a-406b-a11d-a2731cbe7005
• OperateActuator ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23c7-47cd-48a2-b072-ed5d256ff83e
• This new feature of the DSL is 60 relevant for the case of wastewater treatment, but could also be 61 useful in other scenarios or application cases. ls-type:: annotation hl-page:: 4 hl-color:: green id:: 642c23de-93bd-42d2-b645-ebe66b891e89
• o define the types of conditions, metrics to monitor, and1 types of actions that cover our language, we relied on our sys-2 tematic literature review (SLR) (Alfonso et al. 2021b), which3 provides a comprehensive and holistic view of the current state4 of the art in IoT adaptation. ls-type:: annotation hl-page:: 5 hl-color:: green id:: 642c23ec-c040-4e3b-82a1-4641af2550fa
• The definition12 and implementation of these rules improves the accuracy of the13 DSL and avoids errors that could occur at run-time. ls-type:: annotation hl-page:: 5 hl-color:: green id:: 642c24eb-bed5-43f4-afea-68599d38fbb2
• well-formedness rules ls-type:: annotation hl-page:: 5 hl-color:: green id:: 642c2511-5b30-46d7-a41a-c7d43de1f7b0
• projectional editors ls-type:: annotation hl-page:: 5 hl-color:: green id:: 642c2558-9c8d-4931-bb9f-7227d5ba7256
• The color of the shape represents the main fluid1 treated by the unit operation ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c2583-4ad4-4049-a96e-7925d2d75d56
• water treatmen ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c2592-6eb5-448c-88a4-7249c43820f4
• sludge treatmen ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c259a-d8f2-4ad2-9c45-c6a5071f9169
• The modeling of rules—functional or adaptive—is per- 37 formed following a textual notation. A condition, a period, and 38 a list of actions must be defined for a rule. ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c2649-fe37-4de7-99ad-a998b3b06615
• period of the rule ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c266f-bfc3-4e4d-9671-b46860d04035
• he actions of a rule are specified in a vertical list that 60 includes the type of action and all the parameters involved in 61 the action. ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c2677-9fe7-4c3c-954f-986087f9de7e
• Other concepts for modeling the IoT system such as nodes, 77 regions, brokers, applications, and containers can be specified 78 using tabular, textual, and tree-view notations ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c26a4-d01a-4aee-9bbc-47f8fa6f1e94
• MAPE-K is a reference model to implement 83 adaptation mechanisms ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c26c1-e1ce-4cbc-b29f-20671cac823f
• operate on a knowledge base ls-type:: annotation hl-page:: 6 hl-color:: green id:: 642c26c7-e12c-48ed-96e2-9b9d82f1ddca
• Infrastructure and QoS — Infrastructure and QoS metrics9(such as bandwidth, CPU usage, and availability) are col-10 lected using kube-state-metrics4 and node-exporter5 (con-11 tainer cluster monitoring tools) ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c26eb-0417-4be3-aa41-307736598f7a
• Sensor data — The data captured by the system’s sensors, such5 as temperature, suspended solids, level, or pressure. These6 metrics are collected using a monitor subscribed to the7 topics where the sensors publish the data. ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2712-8ca7-4087-a921-dd051b821c32
• Prometheus ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2746-9497-4676-8cb2-bb439a0a2b22
• facilitate the tasks performed in the later stages ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c274c-1a11-4c22-a0d0-67c543904c0a
• analysis ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c274e-b81b-44b6-aaf2-9a68a9506671
• planning ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2750-7a89-4d7c-8453-dd98881c719e
• notification received in the Plan stage, an adapta-32 tion plan is designed with the set of actions—e.g. scaling an33 application— ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2777-e072-4058-9ab8-ad2910d82b1a
• JSON format via HTTP POST requests to37 the Adaptation Engine. ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2783-e185-4f4a-93ab-9ee4332318c3
• K3S6, a Kubernetes-based orches-42 trator optimized for IoT and edge environments ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c288b-852b-4134-800d-da51f36b04df
• ontainer-based IoT applications specified in the input 54 model. ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c2918-3b91-4d46-aabf-5c325b950610
• monitoring tools ls-type:: annotation hl-page:: 8 hl-color:: green id:: 642c291c-04fc-438d-85dc-3a4d8c9acace
• Adaptation Engine. ls-type:: annotation hl-page:: 8 hl-color:: purple id:: 642c2927-6923-457e-934f-b2ac970d1a4a
• DSL for IoT system modeling of a WWTP 70 extends the aspects designed by the language presented in ls-type:: annotation hl-page:: 9 hl-color:: yellow id:: 642c2a85-7e70-43ed-9eeb-4548edbba507
• process block diagram and conditions and actions of a rule10 that involves groups of sensors and actuators ls-type:: annotation hl-page:: 10 hl-color:: green id:: 642c2aa5-1de8-4173-bf1b-2c62a3a374e8