Using Structural Similarity to Predict Future Workload Behavior in the Cloud

Nwanganga, Frederick; Chawla, Nitesh V.

doi:10.1109/cloud.2019.00032

Cited by 4 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiple transactions within a DLRA have strong dependencies across multiple microservices. However, temporal-spatio load variability manifests over time and across nodes [28][29][30] [31]. A user request (e.g., an application request, a database query, a file access operation) will transverse a collection of microservices before being responded.…”

Section: Distributed Long-running Applications (Dlras)mentioning

confidence: 99%

QoS-Aware Co-Scheduling for Distributed Long-Running Applications on Shared Clusters

Zhu

Yang

Sun

et al. 2022

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

To achieve a high degree of resource utilization, production clusters need to co-schedule diverse workloads -including both batch analytic jobs with short-lived tasks and long-running applications (LRAs) that execute for a long time frame from hours to months -onto the shared resources. Microservice architecture advances the manifestation of distributed LRAs (DLRAs), comprising multiple interconnected microservices that are executed in long-lived distributed containers and serve massive user requests. Detecting and mitigating QoS violation become even more intractable due to the network uncertainties and latency propagation across dependent microservices. However, current resource managers are only responsible for resource allocation among applications/jobs but agnostic to runtime QoS such as latency at application level. The state-of-the-art QoS-aware scheduling approaches are dedicated for monolithic applications, without considering the temporal-spatio performance variability across distributed microservices. In this paper, we present TOPOSCH, a new scheduling and execution framework to prioritize the QoS of DLRAs whilst balancing the performance of batch jobs and maintaining high cluster utilization through harvesting idle resources. TOPOSCH tracks footprints of every single request across microservices and uses critical path analysis, based on the end-to-end latency graph, to identify microservices that have high risk of QoS violation. Based on microservice and node level risk assessment, we intervene the batch scheduling by adaptively reducing the visible resources to batch tasks and thus delaying their execution to give way to DLRAs. We propose a prediction-based vertical resource auto-scaling mechanism, with the aid of resource-performance modeling and fine-grained resource inference and access control, for prompt recovery of QoS violation. A cost-effective task preemption is leveraged to ensure a low-cost task preemption and resource reclamation during the auto-scaling. TOPOSCH is integrated with Apache YARN and experiments show that TOPOSCH outperforms other baselines in terms of performance guarantee of DLRAs, at an acceptable cost of batch job slowdown. The tail latency of DLRAs is merely 1.12x of the case of executing alone on average in TOPOSCH with a 26% JCT increase of Spark analytic jobs.

show abstract

Section: Distributed Long-running Applications (Dlras)mentioning

confidence: 99%

QoS-Aware Co-Scheduling for Distributed Long-Running Applications on Shared Clusters

Zhu

Yang

Sun

et al. 2022

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…However, they took into consideration the dynamic opt-in and out of IoT devices into the network, while ignoring their instantaneous workload generation. To the best of our knowledge, the only attempts to estimate workload are referring to the cloud utilization [15,16], and as such they did not capture the locality of the heterogeneous IoT traffics.…”

Section: Iot Workload Profilementioning

confidence: 99%

Edge Computing Resource Allocation for Dynamic Networks: The DRUID-NET Vision and Perspective

Dechouniotis

Αθανασόπουλος

Leivadeas

et al. 2020

Sensors

View full text Add to dashboard Cite

The potential offered by the abundance of sensors, actuators, and communications in the Internet of Things (IoT) era is hindered by the limited computational capacity of local nodes. Several key challenges should be addressed to optimally and jointly exploit the network, computing, and storage resources, guaranteeing at the same time feasibility for time-critical and mission-critical tasks. We propose the DRUID-NET framework to take upon these challenges by dynamically distributing resources when the demand is rapidly varying. It includes analytic dynamical modeling of the resources, offered workload, and networking environment, incorporating phenomena typically met in wireless communications and mobile edge computing, together with new estimators of time-varying profiles. Building on this framework, we aim to develop novel resource allocation mechanisms that explicitly include service differentiation and context-awareness, being capable of guaranteeing well-defined Quality of Service (QoS) metrics. DRUID-NET goes beyond the state of the art in the design of control algorithms by incorporating resource allocation mechanisms to the decision strategy itself. To achieve these breakthroughs, we combine tools from Automata and Graph theory, Machine Learning, Modern Control Theory, and Network Theory. DRUID-NET constitutes the first truly holistic, multidisciplinary approach that extends recent, albeit fragmented results from all aforementioned fields, thus bridging the gap between efforts of different communities.

show abstract

“…Multiple transactions within a DLRA have strong dependencies across multiple microservices. Load variability, however, indicates a temporal-spatio behaviors over time and across nodes [1] [2][3] [4]. A user request (e.g., an application request, a database query, a file access operation) will transverse a collection of microservices before being responded.…”

Section: Renyu Yang Is the Corresponding Authormentioning

confidence: 99%

TOPOSCH: Latency-Aware Scheduling Based on Critical Path Analysis on Shared YARN Clusters

Zhu

Yang

et al. 2020

2020 IEEE 13th International Conference on Cloud Computing (CLOUD)

View full text Add to dashboard Cite

Using Structural Similarity to Predict Future Workload Behavior in the Cloud

Cited by 4 publications

References 21 publications

QoS-Aware Co-Scheduling for Distributed Long-Running Applications on Shared Clusters

QoS-Aware Co-Scheduling for Distributed Long-Running Applications on Shared Clusters

Edge Computing Resource Allocation for Dynamic Networks: The DRUID-NET Vision and Perspective

TOPOSCH: Latency-Aware Scheduling Based on Critical Path Analysis on Shared YARN Clusters

Contact Info

Product

Resources

About