| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | August 4, 2018 |
| Latest Amendment Date: | August 4, 2018 |
| Award Number: | 1763548 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Erik Brunvand
CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2018 |
| End Date: | September 30, 2020 (Estimated) |
| Total Intended Award Amount: | $145,318.00 |
| Total Awarded Amount to Date: | $145,318.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
1918 F ST NW WASHINGTON DC US 20052-0042 (202)994-0728 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
2121 Eye St Washington DC US 20052-0001 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | CSR-Computer Systems Research |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
Smart wearables, the Internet of Things, and new application types, such as augmented reality promise to revolutionize how people interact with technology in their daily lives. While embedded and smart devices have growing capabilities, they still rely on a backend cloud infrastructure to provide additional storage and computational capacity. However, these new application types have characteristics such as strict performance requirements and frequent mobility that are ill suited for today's centralized clouds. This project will develop new system architectures that will increase the scalability, elasticity, and mobility of "edge" applications that connect to mobile users.
Towards this end, the project will explore the communication and system architectures needed to effectively support edge cloud services. The project will leverage advances in network function virtualization to provide high performance networking, and will explore the communication and Operating System primitives needed to support scalable middleboxes and application endpoints. Using this platform as a base, the project will design models that capture the new challenges inherent in mobile edge cloud workloads. These models will be used to guide elastic scaling algorithms.
We are increasingly reliant on mobile computing devices to guide our cars, help us keep in touch with others, gather data of our surroundings, and more. The mobile elastic edge cloud platform being developed in this project will help improve the scalability, agility, and efficiency of edge clouds, allowing them to support new types of performance critical applications. The researchers will engage a broad range of students from the undergraduate to Ph.D. levels in the educational and research activities of this grant.
There will be a project website (http://faculty.cs.gwu.edu/timwood/projects/me2c) that includes all of the artifacts produced throughout the project as well as links to key related technologies and papers. The web repository will include all of the source code developed during the course of the project, documentation with guidance to adopters on using the software, and links to all the papers published and technical reports that are released publicly. The project web page will be maintained for a period of five years after the end of the project.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
Smart wearables, the Internet of Things, and new application types such as augmented reality promise to revolutionize how people interact with technology in their daily lives. These new application types each have unique characteristics such as strict latency requirements and frequent mobility that are ill suited for today's centralized clouds. At the same time, the boundary between the cloud as the computing substrate and the wireless network as the communication substrate is beginning to blur and will soon disappear entirely. In this project, our goal is to design a new Mobile Elastic Edge Cloud (ME2C) architecture that will provide a programmable, converged architecture that can be customized for the diverse needs of emerging applications.
Towards this end, our research has explored the resource management and reliability challenges faced by two critical Edge Cloud technologies: Serverless Computing and Network Function Virtualization (NFV). We have found that while these technologies are relatively well developed for centralized cloud platforms, they have significant limitations for edge cloud environments. In particular, we have see that both Serverless and Network Function performance can dramatically drop when stronger reliability features are added. This is an important concern since many edge- based applications (e.g., cyber physical systems) will be offloading work to the cloud and need strong guarantees that the work will be completed successfully in a timely manner.
The architecture of existing Serverless platforms are highly inefficient–our results show their throughput can be 5-10X slower than un-optimized, “serverfull” web application deployments, while consuming 5X the resources. These overheads arise from design choices that result in repeated request queuing and inefficient communication between the components that make up the Serverless platform. Similarly, NFV platforms have been designed for high performance, but often at the expense of high resource cost. This places them at a severe disadvantage when deployed in an Edge environment where resources are constrained.
Our research has proposed new techniques for improving the reliability and resource efficiency of both NFV and Serverless platforms. We have proposed novel techniques for managing Network Function state that allow efficient, asynchronous backups, while giving guarantees that match more expensive synchronous replication. For Serverless platforms, we propose integrated resource management techniques that holistically control autoscaling, function placement, and request load balancing. Our resource management system provides novel approaches for quickly adapting based on performance feedback metrics that are efficiently propagated with request responses.
The combination of our optimized Serverless platform and our reliable, efficient NFV framework provides the basis for a Mobile Elastic Edge Cloud.
Last Modified: 01/28/2021
Modified by: Timothy Wood
Please report errors in award information by writing to: awardsearch@nsf.gov.
