logseq/pages/hls__Dautov e Distefano - 2022 - Stream Processing on Clustered Edge Devices.md

file:: [Dautov e Distefano - 2022 - Stream Processing on Clustered Edge Devices.pdf](file://C:\Users\david\Zotero/storage/YSV5LG7T/Dautov e Distefano - 2022 - Stream Processing on Clustered Edge Devices.pdf) file-path:: file://C:\Users\david\Zotero/storage/YSV5LG7T/Dautov e Distefano - 2022 - Stream Processing on Clustered Edge Devices.pdf

• network latency and limited bandwidth, this vertical offloading model, however, fails to meet requirements of time-critical data-intensive applications which must act upon generated data with minimum time delays ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bf02-91ab-437b-8f09-301e1862a8d2
• distributed architecture enabling stream data processing at the edge of the network, broadening the principle of enabling processing closer to data sources adopted by Fog and Edge Computing ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bf10-b38e-4d15-861c-5697ad01e30d
• ime-critical IoT applications and services demand for near real-time data processing and reaction, they cannot rely on (potentially outdated) results obtained by sending data over the network to a remote processing location. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bf5b-bb7f-4899-ade1-94ce287e6e9a
• Edge Computing paradigm aims at pushing intelligence to devices that not only provide sensing and actuation resources, but also act as computational nodes in their own right. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bf83-45dc-4b3f-b136-b724e1fd0a63
• Stream Processing architecture to enable horizontal offloading at the edge, ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bf99-a4a9-4aa5-9cd4-80baabbc57a8
• Clustered Edge Computing ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8bfb0-bd3c-4726-b0cc-76c682e45e35
• proposed architecture aims to minimize the amount of data sent to a remote server, reduce network latency, and thus achieve faster processing results. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 63f8c7be-c364-45ce-8b67-24142f7aa230
• For example, the initial object detection can take place immediately at the source, whereas more complex operations are undertaken on a remote server ls-type:: annotation hl-page:: 2 hl-color:: green id:: 63f8c827-2eac-4265-9d81-e23ee2e5ea7b
• resource allocation problem for optimal placement of video analytics queries in such a hierarchy is formulated ls-type:: annotation hl-page:: 2 hl-color:: green id:: 63f8c856-aedc-4794-93e8-c6dc9d74e7a6
• number of hops–a limitation ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 63f8c85f-f22c-4889-999e-c3b9dd311f17
• lack of support for pooling computing resources of multiple collocated edge devices, which only became possible with the recent advances in hardware and networking technologies. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 63f8c894-6908-4d57-ad38-ba2fea82e042
• edge devices can be clustered and managed through middleware at run-time, thereby achieving even lower latency. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 63f8c8a2-5c3c-421e-ae46-18cd1e8682f9
• Similar to the Cloud- and Fog-level coordination, these approaches rely on equipping edge nodes with agent-like virtual containers to enable orchestration and management ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 63f8c8b3-a5a0-4fbd-b9b3-a8118e547a06
• edge devices are able to communicate with each other to split, delegate, and share processing tasks. ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 63f8c8bc-8a8b-46f3-9815-2797919abfda
• cloudlets ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 63f8c8e4-64ec-47ac-a0ce-6b6d36bef4ed
• cluster initiators/coordinators and worker nodes ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 63f8c904-e251-4a6a-803d-497789db8302
• dynamic discovery, selection and management of suitable nodes at run-time. ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 63f8c911-91eb-4688-82ba-2590ae161dac
• tream Processing middleware for in-memory data analytics ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 63f8c926-cb6e-4a19-92cf-b4bda9d58a68
• dynamic clustering and task offloading at system run-time ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 63f8c932-52d7-42b5-859b-d7a816d8211d
• dge Computing has to be enhanced with clustering techniques extending its application domain towards Clustered Edge Computing (CEC) ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 63f8c95a-e7c0-4d93-812b-d578335f707f