<i>AutoBoT</i>: Resilient and Cost-Effective Scheduling of a Bag of Tasks on Spot VMs

Varshney, Prateeksha; Simmhan, Yogesh

doi:10.1109/tpds.2018.2889851

Cited by 28 publications

(26 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A bag of tasks (BoT) consists of a homogeneous batch of tasks that can be executed in any order. BoTs can achieve a high degree of task parallelism and efficiency across VMs and devices since they do not have interdependencies …”

Section: The Taxonomymentioning

confidence: 99%

“…They use heuristics such as predicting the proximity between mobile edge devices for deadline planning, prioritizing tasks with the most resource constraints, and incrementally colocating tasks on the same resource to avoid network latency. More sophisticated heuristics have also been proposed, such as the ToF planner and AutoBoT that uses a greedy approach to minimize the monetary cost for executing a bag of workflows or tasks on clouds . Similarly, backtracking algorithms follow one of many possible alternatives and backtrack if it does not look promising …”

Section: The Taxonomymentioning

confidence: 99%

“…Random DAG generators are also used to generate DAGs with parameters and distributions to control the number of tasks, number of dependency edges, range of task length, depth and width of DAG, outdegree, and communication to computation ratio . Alternatively, publicly available cloud workloads, such as the Google Cluster Workload also offer the choice of realistic DAG and BoT simulations . The CloudSim and iFogSim simulation frameworks can model application execution on cloud infrastructure and edge and fog infrastructure, respectively, and are often used to evaluate scheduling algorithms .…”

Section: The Taxonomymentioning

confidence: 99%

“…BoTs can achieve a high degree of task parallelism and efficiency across VMs and devices since they do not have interdependencies. 113,121,125 A transactional mode is common when multiple users share the same application deployed on a resource (eg, function as a service (FAAS)), or a stream of microbatches arrive from an input source for processing by the application. Since requests arrive continuously, future workloads are hard to predict, and resources need to be dynamically allocated and deallocated.…”

Section: Mode Of Submissionmentioning

confidence: 99%

“…More sophisticated heuristics have also been proposed, such as the ToF planner and AutoBoT that uses a greedy approach to minimize the monetary cost for executing a bag of workflows or tasks on clouds. 101,121 Similarly, backtracking algorithms follow one of many possible alternatives and backtrack if it does not look promising. 66,117 Workflow scheduling problems are often reduced to integer linear programming (ILP) and its mixed variant.…”

Section: Scheduling Techniquesmentioning

confidence: 99%

See 4 more Smart Citations

Characterizing application scheduling on edge, fog, and cloud computing resources

Varshney

Simmhan

2019

Softw Pract Exp

Self Cite

View full text Add to dashboard Cite

Summary Cloud computing has grown to become a popular distributed computing service offered by commercial providers. More recently, edge and fog computing resources have emerged on the wide‐area network as part of Internet of things (IoT) deployments. These three resource abstraction layers are complementary, and offer distinctive benefits. Scheduling applications on clouds has been an active area of research, with workflow and data flow models offering a flexible abstraction to specify applications for execution. However, the application programming and scheduling models for edge and fog are still maturing, and can benefit from learnings on cloud resources. At the same time, there is also value in using these resources cohesively for application execution. In this article, we offer a taxonomy of concepts essential for specifying and solving the problem of scheduling applications on edge, fog, and cloud computing resources. We first characterize the resource capabilities and limitations of these infrastructure and offer a taxonomy of application models, quality‐of‐service constraints and goals, and scheduling techniques, based on a literature review. We also tabulate key research prototypes and papers using this taxonomy. This survey benefits developers and researchers on these distributed resources in designing and categorizing their applications, selecting the relevant computing abstraction(s), and developing or selecting the appropriate scheduling algorithm. It also highlights gaps in literature where open problems remain.

show abstract

Section: The Taxonomymentioning

confidence: 99%

Section: The Taxonomymentioning

confidence: 99%