skip to main content

Attention:

The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Thursday, June 13 until 2:00 AM ET on Friday, June 14 due to maintenance. We apologize for the inconvenience.


Title: Dělen: Enabling Flexible and Adaptive Model-serving for Multi-tenant Edge AI
Model-serving systems expose machine learning (ML) models to applications programmatically via a high-level API. Cloud plat- forms use these systems to mask the complexities of optimally managing resources and servicing inference requests across multi- ple applications. Model serving at the edge is now also becoming increasingly important to support inference workloads with tight latency requirements. However, edge model serving differs substan- tially from cloud model serving in its latency, energy, and accuracy constraints: these systems must support multiple applications with widely different latency and accuracy requirements on embedded edge accelerators with limited computational and energy resources. To address the problem, this paper presents Dělen,1 a flexible and adaptive model-serving system for multi-tenant edge AI. Dělen exposes a high-level API that enables individual edge applications to specify a bound at runtime on the latency, accuracy, or energy of their inference requests. We efficiently implement Dělen using conditional execution in multi-exit deep neural networks (DNNs), which enables granular control over inference requests, and evalu- ate it on a resource-constrained Jetson Nano edge accelerator. We evaluate Dělen flexibility by implementing state-of-the-art adapta- tion policies using Dělen’s API, and evaluate its adaptability under different workload dynamics and goals when running single and multiple applications.  more » « less
Award ID(s):
2213636 2211888 2105494 2211302
NSF-PAR ID:
10433378
Author(s) / Creator(s):
; ; ; ; ;
Date Published:
Journal Name:
ACM/IEEE Conference on Internet of Things Design and Implementation
Page Range / eLocation ID:
209 to 221
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Edge cloud data centers (Edge) are deployed to provide responsive services to the end-users. Edge can host more powerful CPUs and DNN accelerators such as GPUs and may be used for offloading tasks from end-user devices that require more significant compute capabilities. But Edge resources may also be limited and must be shared across multiple applications that process requests concurrently from several clients. However, multiplexing GPUs across applications is challenging. With edge cloud servers needing to process a lot of streaming and the advent of multi-GPU systems, getting that data from the network to the GPU can be a bottleneck, limiting the amount of work the GPU cluster can do. The lack of prompt notification of job completion from the GPU can also result in poor GPU utilization. We build on our recent work on controlled spatial sharing of a single GPU to expand to support multi-GPU systems and propose a framework that addresses these challenges. Unlike the state-of-the-art uncontrolled spatial sharing currently available with systems such as CUDA-MPS, our controlled spatial sharing approach uses each of the GPU in the cluster efficiently by removing interference between applications, resulting in much better, predictable, inference latency We also use each of the cluster GPU's DMA engines to offload data transfers to the GPU complex, thereby preventing the CPU from being the bottleneck. Finally, our framework uses the CUDA event library to give timely, low overhead GPU notifications. Our evaluations show we can achieve low DNN inference latency and improve DNN inference throughput by at least a factor of 2. 
    more » « less
  2. Smart IoT-based systems often desire continuous execution of multiple latency-sensitive Deep Learning (DL) appli- cations. The edge servers serve as the cornerstone of such IoT- based systems, however, their resource limitations hamper the continuous execution of multiple (multi-tenant) DL applications. The challenge is that, DL applications function based on bulky “neural network (NN) models” that cannot be simultaneously maintained in the limited memory space of the edge. Accordingly, the main contribution of this research is to overcome the memory contention challenge, thereby, meeting the latency constraints of the DL applications without compromising their inference accuracy. We propose an efficient NN model management frame- work, called Edge-MultiAI, that ushers the NN models of the DL applications into the edge memory such that the degree of multi-tenancy and the number of warm-starts are maximized. Edge-MultiAI leverages NN model compression techniques, such as model quantization, and dynamically loads NN models for DL applications to stimulate multi-tenancy on the edge server. We also devise a model management heuristic for Edge-MultiAI, called iWS-BFE, that functions based on the Bayesian theory to predict the inference requests for multi-tenant applications, and uses it to choose the appropriate NN models for loading, hence, increasing the number of warm-start inferences. We evaluate the efficacy and robustness of Edge-MultiAI under various configurations. The results reveal that Edge-MultiAI can stimulate the degree of multi-tenancy on the edge by at least 2× and increase the number of warm-starts by ≈ 60% without any major loss on the inference accuracy of the applications. 
    more » « less
  3. The dramatic growth in demand for mobile data service has prompted mobile network operators (MNOs) to explore new spectrum resources in unlicensed bands. MNOs have been recently allowed to extend LTE-based service called LTE-LAA over 5 GHz U-NII bands, currently occupied by Wi-Fi. To support applications with diverse QoS requirements, both LTE and Wi-Fi technologies introduce multiple priority classes with different channel contention parameters for accessing unlicensed bands. How these different priority classes affect the interplay between coexisting LTE and Wi-Fi technologies is still relatively under-explored. In this paper, we develop a simple and efficient framework that helps MNOs assess the fair coexistence between MNOs and Wi-Fi operators with prioritized channel access under the multi-channel setting. We derive an approximated closed-form solution for each MNO to pre-evaluate the probability of successful transmission (PST), average contention delay, and average throughput when adopting different priority classes to serve different traffics. MNOs and Wi-Fi operators can fit our model using measurements collected offline and/or online, and use it to further optimize their systems’ throughput and latency. Our results reveal that PSTs computed with our approximated closed-form model approach those collected from system-level simulations with around 95% accuracy under scenarios of dense network deployment density and high traffic intensity. 
    more » « less
  4. null (Ed.)
    Internet of Things (IoT) devices are becoming increasingly prevalent in our environment, yet the process of programming these devices and processing the data they produce remains difficult. Typically, data is processed on device, involving arduous work in low level languages, or data is moved to the cloud, where abundant resources are available for Functions as a Service (FaaS) or other handlers. FaaS is an emerging category of flexible computing services, where developers deploy self-contained functions to be run in portable and secure containerized environments; however, at the moment, these functions are limited to running in the cloud or in some cases at the "edge" of the network using resource rich, Linux-based systems. In this work, we investigate NanoLambda, a portable platform that brings FaaS, high-level language programming, and familiar cloud service APIs to non-Linux and microcontroller-based IoT devices. To enable this, NanoLambda couples a new, minimal Python runtime system that we have designed for the least capable end of the IoT device spectrum, with API compatibility for AWS Lambda and S3. NanoLambda transfers functions between IoT devices (sensors, edge, cloud), providing power and latency savings while retaining the programmer productivity benefits of high-level languages and FaaS. A key feature of NanoLambda is a scheduler that intelligently places function executions across multi-scale IoT deployments according to resource availability and power constraints. We evaluate a range of applications that use NanoLambda to run on devices as small as the ESP8266 with 64KB of ram and 512KB flash storage. 
    more » « less
  5. Deep neural network (DNN) inference poses unique challenges in serving computational requests due to high request intensity, concurrent multi-user scenarios, and diverse heterogeneous service types. Simultaneously, mobile and edge devices provide users with enhanced computational capabilities, enabling them to utilize local resources for deep inference processing. Moreover, dynamic inference techniques allow content-based computational cost selection per request. This paper presents Dystri, an innovative framework devised to facilitate dynamic inference on distributed edge infrastructure, thereby accommodating multiple heterogeneous users. Dystri offers a broad applicability in practical environments, encompassing heterogeneous device types, DNN-based applications, and dynamic inference techniques, surpassing the state-of-the-art (SOTA) approaches. With distributed controllers and a global coordinator, Dystri allows per-request, per-user adjustments of quality-of-service, ensuring instantaneous, flexible, and discrete control. The decoupled workflows in Dystri naturally support user heterogeneity and scalability, addressing crucial aspects overlooked by existing SOTA works. Our evaluation involves three multi-user, heterogeneous DNN inference service platforms deployed on distributed edge infrastructure, encompassing seven DNN applications. Results show Dystri achieves near-zero deadline misses and excels in adapting to varying user numbers and request intensities. Dystri outperforms baselines with accuracy improvement up to 95 ×. 
    more » « less