{"corpus_id":797368,"paper_sha":"1ebd47c89045f7a71eb304fb18cfd1c33e591047","doi":"10.1109/IC2E.2016.36","arxiv_id":"1604.04804","pmid":null,"pmcid":null,"mag_id":2338690345,"dblp_id":"conf/ic2e/DoyleGAA16","acl_id":null,"title":"Cloud Instance Management and Resource Prediction for Computation-as-a-Service Platforms","year":2016,"publication_date":"2016-04-16","venue":"2016 IEEE International Conference on Cloud Engineering (IC2E)","journal":{"name":"2016 IEEE International Conference on Cloud Engineering (IC2E)","pages":"89-98","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle","Conference"],"pubmed_pub_types":null,"s2_fields_of_study":["Computer Science","Engineering"],"reference_count":32,"citation_count":10,"influential_citation_count":0,"is_open_access":true,"arxiv_categories":["cs.DC"],"arxiv_license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":"http://arxiv.org/pdf/1604.04804","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/1ebd47c89045f7a71eb304fb18cfd1c33e591047","s2_open_access_license":null,"s2_open_access_status":"GREEN","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Computation-as-a-Service (CaaS) offerings have gained traction in the last few years due to their effectiveness in balancing between the scalability of Software-as-a-Service and the customisation possibilities of Infrastructure-as-a-Service platforms. To function effectively, a CaaS platform must have three key properties: (i) reactive assignment of individual processing tasks to available cloud instances (compute units) according to availability and predetermined time-to-completion (TTC) constraints, (ii) accurate resource prediction, (iii) efficient control of the number of cloud instances servicing workloads, in order to optimize between completing workloads in a timely fashion and reducing resource utilization costs. In this paper, we propose three approaches that satisfy these properties (respectively): (i) a service rate allocation mechanism based on proportional fairness and TTC constraints, (ii) Kalman-filter estimates for resource prediction, and (iii) the use of additive increase multiplicative decrease (AIMD) algorithms (famous for being the resource management in the transport control protocol) for the control of the number of compute units servicing workloads. The integration of our three proposals into a single CaaS platform is shown to provide for more than 27% reduction in Amazon EC2 spot instance cost against methods based on reactive resource prediction and 38% to 60% reduction of the billing cost against the current state-of-the-art in CaaS platforms (Amazon Lambda and Autoscale).","claims":[{"public_id":"cl_7a2a568fd3f32953db39aa7112fa2395","status":"active","text":"A service rate allocation mechanism based on proportional fairness and time-to-completion constraints satisfies reactive assignment requirements for cloud instances.","confidence":0.96,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_7a2a568fd3f32953db39aa7112fa2395"},{"public_id":"cl_8e17230991dce72db77b8213eb17a2f1","status":"active","text":"Additive increase multiplicative decrease algorithms are used to control the number of compute units servicing workloads.","confidence":0.96,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_8e17230991dce72db77b8213eb17a2f1"},{"public_id":"cl_2864e097c52f576a0678b6d84a030b05","status":"active","text":"Integrating the three proposals into a single computation-as-a-service platform reduces Amazon EC2 spot instance cost by more than 27% relative to reactive resource prediction methods and reduces billing cost by 38% to 60% relative to Amazon Lambda and Autoscale.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_2864e097c52f576a0678b6d84a030b05"},{"public_id":"cl_541412be204c41d22adaa554251e2076","status":"active","text":"Kalman-filter estimates are used for resource prediction in the proposed computation-as-a-service platform.","confidence":0.95,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_541412be204c41d22adaa554251e2076"}],"concepts":[{"public_id":"co_00b6a81aa370788f49a3e639f8640f56","status":"active","name":"time-to-completion constraints","description":"Predetermined limits on how long tasks are allowed to take before completion.","types":["constraint"],"aliases":["TTC constraints"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_00b6a81aa370788f49a3e639f8640f56"},{"public_id":"co_0dc02cdb4a9d23e7f766e4e37b91aeaa","status":"active","name":"workloads","description":"Sets of processing tasks that need to be completed on the platform.","types":["task set"],"aliases":["processing tasks"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_0dc02cdb4a9d23e7f766e4e37b91aeaa"},{"public_id":"co_11c043a2b0e4714e021d6a3220ab720e","status":"active","name":"additive increase multiplicative decrease algorithms","description":"Feedback control algorithms that raise a quantity gradually and reduce it multiplicatively when needed.","types":["control algorithm"],"aliases":["AIMD algorithms"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_11c043a2b0e4714e021d6a3220ab720e"},{"public_id":"co_28127aed5607867537c0e3aa8f6120d3","status":"active","name":"compute units","description":"Cloud instances that provide processing capacity for servicing workloads.","types":["resource"],"aliases":["cloud instances"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_28127aed5607867537c0e3aa8f6120d3"},{"public_id":"co_48b5be2ba87ec5f0d207b2898caff180","status":"active","name":"Autoscale","description":"An autoscaling system used as a comparison baseline for billing-cost evaluation.","types":["baseline"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_48b5be2ba87ec5f0d207b2898caff180"},{"public_id":"co_610ef4cfc3e07cda598b64a3410701c3","status":"active","name":"Amazon Lambda","description":"A managed serverless compute service used as a state-of-the-art comparison system.","types":["baseline"],"aliases":["AWS Lambda"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_610ef4cfc3e07cda598b64a3410701c3"},{"public_id":"co_74738556baa8d922c6c9f75b95012f91","status":"active","name":"Amazon EC2 spot instance cost","description":"The cost of using Amazon EC2 spot instances for cloud computation.","types":["cost metric"],"aliases":["EC2 spot instance cost"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_74738556baa8d922c6c9f75b95012f91"},{"public_id":"co_8d8871df00b62f1a68f099c3ff99bc16","status":"active","name":"proportional fairness","description":"An allocation principle that balances utility across competing tasks or users.","types":["allocation principle"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8d8871df00b62f1a68f099c3ff99bc16"},{"public_id":"co_8d89f1e3ae0bcde75af8381a1f848c8e","status":"active","name":"computation-as-a-service platform","description":"A cloud platform that provides computation services while balancing scalability and customization.","types":["platform"],"aliases":["CaaS platform"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8d89f1e3ae0bcde75af8381a1f848c8e"},{"public_id":"co_9e9df3cabf70ff1a3f5002c231a4a477","status":"active","name":"resource prediction","description":"Estimation of future resource requirements for workloads running on cloud infrastructure.","types":["method"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_9e9df3cabf70ff1a3f5002c231a4a477"},{"public_id":"co_cb0b60e944b3655c1337e9ceef75301c","status":"active","name":"service rate allocation mechanism","description":"A mechanism for assigning processing service rates to cloud instances or tasks.","types":["method"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_cb0b60e944b3655c1337e9ceef75301c"},{"public_id":"co_eaf212f0623d012f22ccab3e6341d7fb","status":"active","name":"Kalman-filter estimates","description":"Predictions produced using a Kalman filter for estimating system state over time.","types":["method"],"aliases":["Kalman filter estimates"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_eaf212f0623d012f22ccab3e6341d7fb"}],"external_ids":{"DOI":"10.1109/IC2E.2016.36","ArXiv":"1604.04804","PubMed":null,"PubMedCentral":null,"MAG":2338690345,"DBLP":"conf/ic2e/DoyleGAA16","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"http://arxiv.org/pdf/1604.04804","landing_url":"https://www.semanticscholar.org/paper/1ebd47c89045f7a71eb304fb18cfd1c33e591047","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":null,"status":"GREEN","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":true,"full_text_source":"arxiv","count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":638325,"paper_uid":"9c8f5183-9272-48ad-a3f4-c4e7847b7a51","canonical_identity":{"paper_id":638325,"paper_uid":"9c8f5183-9272-48ad-a3f4-c4e7847b7a51","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/797368"}