The Fourth ACM/IEEE Symposium on Edge Computing (SEC) accepted a paper by Professor Rich Wolski & Chandra Krintz.

Edge Computing is a new computing paradigm where server resources, ranging from a credit-card size computer to a small data center, are placed closer to data and information generation sources. Application and systems developers use these resources to enable a new class of latency and bandwidth sensitive applications that are not realizable with current cloud computing architectures. Edge computing represents a counterpoint to the consolidation of computing into massive data centers, which has dominated the discourse in cloud computing for well over a decade. Popular terms such as micro-data centers, intelligent edges, cloudlets, and fog have been used interchangeably to describe edge computing.

 

CSPOT: Portable, Multi-scale Functions-as-a-Service for IoT

Rich Wolski, Chandra Krintz, Fatih Bakir, Gareth George, and Wei-Tsung Lin {rich,ckrintz,bakir,gareth,weitsung}@cs.ucsb.edu
Computer Science Dept.
Univ. of California, Santa Barbara 

In this paper, we present CSPOT, a distributed runtime system implementing a functions-as-service (FaaS) programming model for the “Internet of Things” (IoT). With FaaS, developers express arbitrary computations as simple functions that are automatically invoked and managed by a cloud platform in response to events. We extend this FaaS model so that it is suitable for use in all tiers of scale for IoT – sensors, edge devices, and cloud – to facilitate robust, portable, and low-latency IoT application development and deployment.

To enable this, we combine the use of Linux containers and namespaces for isolation and portability, an append-only, object store for robust persistence, and a causal event log for triggering functions and tracking event dependencies. We present the design and implementation of CSPOT, detail its abstractions and APIs, and overview examples of its use. We empirically evaluate the performance of CSPOT using different devices and applications and find that it implements function invocation with significantly lower latency than other FaaS offerings, while providing portability across tiers and similar data durability characteristics.