BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:Asia/Kolkata X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT UID:84@cds.iisc.ac.in DTSTART;TZID=Asia/Kolkata:20241204T103000 DTEND;TZID=Asia/Kolkata:20241204T113000 DTSTAMP:20241128T173830Z URL:https://cds.iisc.ac.in/events/ph-d-thesis-defense-cds-december-04-2024 -designing-quality-of-service-aware-serverless-platforms/ SUMMARY:Ph.D. Thesis Defense: CDS: December 04\, 2024 "Designing Quality of Service aware Serverless Platforms" DESCRIPTION:DEPARTMENT OF COMPUTATIONAL AND DATA SCIENCES\nPh.D. Thesis Def ense\n\n\n\nSpeaker : Mr. Sheshadri Kalkunte Ramachandra\nS.R. Number : 06 -18-02-10-12-18-1-16336\nTitle : "Designing Quality of Service aware Serve rless Platforms"\nResearch Supervisor: Dr. J. Lakshmi\nThesis examiner : D r. Nitin Auluck\, IIT-Ropar\nDate &\; Time : December 04\, 2024 (Wednes day) at 10:30 AM\nVenue : The Thesis Colloquium will be held on HYBRID Mod e\n# 102 CDS Seminar Hall /MICROSOFT TEAMS.\nPlease click on the following link to join the Thesis Colloquium:\nMS Teams link\n\n\n\nABSTRACT\n\nSer verless computing is a widely used Cloud computing service offering that p rovides users with managed runtimes to develop their business logic as fun ctions. It supports event-driven execution of functions and a powerful “ pay-as-you-use” billing model that\nbills the functions based on the dur ation of execution for the resources assigned. The service provider throug h the serverless platform performs application deployment\, resource manag ement\, provisioning\, and system administration tasks. Though this automa tic\nresource management results in ease of use for the user\, the platfor m internally performs many interactions to orchestrate the function execut ion\, leading to significant overheads. Platform overheads coupled with th e prevalence of hardware heterogeneity in\nthe Cloud and resource specific ation that is agnostic to application characteristics cause non-determinis m in function executions. As a result applications do not have any guarant ees from the platform to meet their expectation\; and existing serverless platforms\nprovide no way for the users to express their application requi rements to the platform.\n\nQuality of Service(QoS) is a generic mechanism for the application to describe its expectation from the execution in a p latform neutral way. The serverless platform can use the QoS specification s to make decisions about resource allocation and provisioning to modulate the application behaviour in accordance to the desired QoS. The challenge in designing for QoS support is its specificity to application domain. Di fferent domains exhibit different requirements and it may not be feasible or realistic to design one QoS model that suits every application. To unde rstand and evaluate the design challenges and impact of QoS on application behaviour on serverless platforms\, the thesis identifies three prevalent application domains which give a fair representation of challenges across different application domains that have adopted serverless computing. In the context of these identified applications\, the thesis proposes how ser verless platforms can be designed to address application requirements whil e making resource provisioning and deployment decisions.\n\nThe first cont ribution explores the domain of Internet of Things. The applications in th is domain are typically deployed across Edge and Cloud layers and have var ying image and video data as input\; they are triggered based on events. P erformance characteristic\, response latency is the primary QoS attribute considered for this domain. This work addresses the challenges of deducing resource specifications and performing function deployment in a heterogen eous resource setting. The proposed solution “QoS aware serverless platf orm for Edge-Cloud continuum” uses application characteristics and QoS r equirements to deploy functions across Edge and Cloud with an objective of meeting user-specified QoS requirements. It also exploits data locality i n function placement to improve function execution latency. Proposed desig n is evaluated using representative application scenario\, and results ana lysed for efficacy\, improvements and pitfalls.\n\nThe second contribution explores the domain of Network Function Virtualization (NFV). NFV allows network functions to be realized as software entities\; they are stitched together with other network functions to form a Network Service or a Servi ce Function Chain (SFC)\, and are deployed on Cloud as an integral part of the infrastructure. SFCs are used to enforce organizational policies on t he incoming and outgoing network traffic\, typically for a tenant. Network traffic is akin to streaming data\; it requires persistent services to ha ve responsive processing of traffic to meet application requirements. In t his work\, end-to-end latency and ordering are the application requirement s. The existing solutions that deploy the SFCs on the Cloud using Infrastr ucture as a Service (IaaS) or Function as a Service (FaaS) suffer from the problems of poor scalability and high platform overheads\, respectively. In order to address these shortcomings\, a Hybrid Serverless Platform (HSP ) is proposed that comprises the best of both worlds\, IaaS and FaaS. The proposed solution uses IaaS for steady state workloads\, and FaaS to servi ce the network workload in phases of scaling IaaS or failures of IaaS inst ances to avoid loss of service. In addition\, a succinct way to specify te nant specific requirements to the platform is also discussed. Experiments based on real-world workloads show the benefits of the Hybrid serverless d esign and QoS-based resource management.\n\nThe third contribution focuses on the domain of Inferencing which constitutes an important class of even t-driven applications. The applications in this domain are designed as a w orkflow of functions that process images and texts as input\; they are typ ically deployed across Edge and Cloud layers. Response Latency is the QoS attribute under consideration. This work addresses the problem of platform overheads in enacting inferencing workflows. Existing solutions suggest f using functions to avoid overheads. However\, they result in high executio n costs and do not address the case of function fusion across heterogeneou s resource layers. To address this\, RightFusion\, a novel fusion criteria that performs QoS aware Function Fusion across Edge and Cloud is proposed \; it uses application characteristics and QoS requirements to perform low -cost function executions that result in significant resource savings comp ared to existing solutions. Results analysis also discuss the usefulness a nd shortcomings of the design to bring out its applicability.\n\nThe work described in the thesis demonstrates the value of considering QoS requirem ents in serverless platforms for both the user and provider. It enables th e users to convey expected or desired application behavior\, thereby bring ing in a sense of guarantee during its execution on serverless platforms. It also enhances the platform's response in consideration of the desired r equirements of the application\, thereby improving provider credibility. Q oS specification also enables smarter resource management in serverless pl atforms\, leading to resource savings for the provider and reduced cost fo r the user. Though QoS varies from one application domain to another\, the requirements can be addressed by bringing in necessary design customizati on to different components in the existing serverless architecture\; these changes are applicable to serverless platforms in general\, irrespective of the application domain. While latency was the primary QoS attribute in our study\, other QoS features\, such as security\, privacy\, and reliabil ity\, present interesting research directions to explore.\n\n\n\nALL ARE W ELCOME CATEGORIES:Events,Thesis Defense END:VEVENT BEGIN:VTIMEZONE TZID:Asia/Kolkata X-LIC-LOCATION:Asia/Kolkata BEGIN:STANDARD DTSTART:20231205T103000 TZOFFSETFROM:+0530 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE END:VCALENDAR