Timely-Throughput Optimal Coded Computing over Cloud Networks

Yang, Chien-Sheng; Pedarsani, Ramtin; Avestimehr, A. Salman

doi:10.1145/3323679.3326528

Cited by 41 publications

(23 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Coded computing is a recent technique that enables optimal tradeoffs between computation load, communication load, and computation latency due to stragglers in distributed computing (see, e.g., [5], [6], [37]- [39]). Therefore, designing joint task scheduling and coded computing in order to leverage tradeoffs between computation, communication, and latency could be an interesting problem (e.g., [40]). µ (k,j) p (k,j) , = r (k,j) − r (k+1,j) (104) = µ (k,j) p (k,j) − µ (k+1,j) p (k+1,j) .…”

Section: Discussionmentioning

confidence: 99%

Communication-Aware Scheduling of Serial Tasks for Dispersed Computing

Yang

Pedarsani

Avestimehr

2019

IEEE/ACM Trans. Networking

Self Cite

View full text Add to dashboard Cite

There is a growing interest in the development of in-network dispersed computing paradigms that leverage the computing capabilities of heterogeneous resources dispersed across the network for processing massive amount of data collected at the edge of the network. We consider the problem of task scheduling for such networks, in a dynamic setting in which arriving computation jobs are modeled as chains, with nodes representing tasks, and edges representing precedence constraints among tasks. In our proposed model, motivated by significant communication costs in dispersed computing environments, the communication times are taken into account. More specifically, we consider a network where servers can serve all task types, and sending the outputs of processed tasks from one server to another server results in some communication delay. We first characterize the capacity region of the network, then propose a novel virtual queueing network encoding the state of the network. Finally, we propose a Max-Weight type scheduling policy, and considering the stochastic network in the fluid limit, we use a Lyapunov argument to show that the policy is throughput-optimal. Beyond the model of chains, we extend the scheduling problem to the model of directed acyclic graph (DAG) which imposes a new challenge, namely logic dependency difficulty, requiring the data of processed parents tasks to be sent to the same server for processing the child task. We propose a virtual queueing network for DAG scheduling over broadcast networks, where servers always broadcast the data of processed tasks to other servers, and prove that Max-Weight policy is throughput-optimal. Now, we focus on job m specified by a chain and compute J j=1 r (k,j) . If k ∈ C, we have J j=1 r (k,j) = J j=1 λ m(k) u m→j = λ m(k) J j=1 u m→j = λ m(k) since J j=1 u m→j = 1. Then, we can compute J j=1 r (k+1,j) as follows: J j=1 r (k+1,j) = J j=1 J l=1 r (k,l) f k,l→j = J l=1 r (k,l) J j=1 f k,l→j = J l=1 r (k,l) = λ m(k) (98) since J j=1 f k,l→j = 1. By induction, we have J j=1 r (k,j) = λ m(k) , ∀k ∈ I m . Then, we have J j=1 r (k,j) = J j=1

show abstract

Section: Discussionmentioning

confidence: 99%

Communication-Aware Scheduling of Serial Tasks for Dispersed Computing

Yang

Pedarsani

Avestimehr

2019

IEEE/ACM Trans. Networking

Self Cite

View full text Add to dashboard Cite

show abstract

“…according to Eq. (7). Using proof of contradiction, we can then derive that the infimum and supremum of λ i are attained when p i → ∞ and p i = 1, respectively.…”

Section: Appendix Proof Of Lemmamentioning

confidence: 96%

“…One major challenge is that many computing frameworks are vulnerable to uncertain disturbances, such as node/link failures, communication congestion, and slow-downs [6]. Such disturbances, which can be modeled as stragglers that are slow or even fail in returning results, have been observed in many largescale computing systems such as cloud computing [7], mobile edge computing [8], and fog computing [9].…”

Section: Introductionmentioning

confidence: 99%

On Batch-Processing Based Coded Computing for Heterogeneous Distributed Computing Systems

Wang

Xie

et al. 2021

IEEE Trans. Netw. Sci. Eng.

View full text Add to dashboard Cite

In recent years, coded distributed computing (CDC) has attracted significant attention, because it can efficiently facilitate many delay-sensitive computation tasks against unexpected latencies in distributed computing systems. Despite such a salient feature, many design challenges and opportunities remain. In this paper, we focus on practical computing systems with heterogeneous computing resources, and design a novel CDC approach, called batch-processing based coded computing (BPCC), which exploits the fact that every computing node can obtain some coded results before it completes the whole task. To this end, we first describe the main idea of the BPCC framework, and then formulate an optimization problem for BPCC to minimize the task completion time by configuring the computation load. Through formal theoretical analyses, extensive simulation studies, and comprehensive real experiments on the Amazon EC2 computing clusters, we demonstrate promising performance of the proposed BPCC scheme, in terms of high computational efficiency and robustness to uncertain disturbances.

show abstract

“…The first work found the exact expression of the average AoI, while the latter found its entire distribution. Timely applications in the edge/cloud were studied in [9]- [12]. Particularly, in the latter, a feedback loop is inserted as a means of reducing the load at the server side, but only the percentage of aborted tasks is considered as a metric of the advantage of having feedback.…”

Section: Introductionmentioning

confidence: 99%

Load Reduction for Timely Applications in the Cloud

et al. 2021

View full text Add to dashboard Cite

In many IoT applications, sensor data is sent remotely to be processed, but only the freshest result is of interest. In this paper we investigate a feedback mechanism that aborts the processing of stale data at the remote end to reduce the load and save costs. The process is approximated by an M/M/∞ queueing system with a feedback loop. We find the exact expression of the average computational time saved and show that with the feedback loop in place the computation time per CPU can be cut up to 25%, making the technique very promising.

show abstract

Timely-Throughput Optimal Coded Computing over Cloud Networks

Cited by 41 publications

References 34 publications

Communication-Aware Scheduling of Serial Tasks for Dispersed Computing

Communication-Aware Scheduling of Serial Tasks for Dispersed Computing

On Batch-Processing Based Coded Computing for Heterogeneous Distributed Computing Systems

Load Reduction for Timely Applications in the Cloud

Contact Info

Product

Resources

About