| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | August 13, 2017 |
| Latest Amendment Date: | September 3, 2021 |
| Award Number: | 1743358 |
| Award Instrument: | Cooperative Agreement |
| Program Manager: |
Deepankar Medhi
dmedhi@nsf.gov (703)292-2935 CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2017 |
| End Date: | September 30, 2022 (Estimated) |
| Total Intended Award Amount: | $5,379,684.00 |
| Total Awarded Amount to Date: | $5,395,684.00 |
| Funds Obligated to Date: |
FY 2018 = $1,432,361.00 FY 2019 = $845,140.00 FY 2020 = $16,000.00 |
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
5801 S ELLIS AVE CHICAGO IL US 60637-5418 (773)702-8669 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
5735 S. Ellis Avenue Chicago IL US 60637-5418 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): |
Special Projects - CNS, CISE Research Resources |
| Primary Program Source: |
01001819DB NSF RESEARCH & RELATED ACTIVIT 01001920DB NSF RESEARCH & RELATED ACTIVIT 01002021DB NSF RESEARCH & RELATED ACTIVIT |
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
One of two NSFCloud mid-scale experimental computing research infrastructures, Chameleon is a deeply reconfigurable cloud computing testbed supporting large-scale Computer Science experimentation. Located at the University of Chicago and the Texas Advanced Computing Center, it is composed of over 600 compute nodes and a total of 5 petabytes of storage supporting data-intensive science. Since its public availability in July 2015, Chameleon has attracted a community of over 1,300 users working on over 200 education and research projects.
In Phase 2 of the NSFCloud program, this project will expand Chameleon capabilities and capacity, and will fulfill its mission of serving as a testbed that is used extensively by the research community by:
1) broadening the set of supported experiments through the addition of new hardware and software supporting high bandwidth and Software Defined Networking (SDN)
2) providing operational sustainability by streamlining operations, packaging the testbed as open source software that can be leveraged by others, and developing a process for adding resources to the testbed
3) provide support for managing, understanding, comparing and reproducing Computer Science experiments, and
4) focusing effort on building an engaged community around the testbed via both traditional outreach including targeted training and education efforts, and direct personalized outreach to experimenters.
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.
Chameleon Outcomes
The Chameleon project designed and deployed a scientific instrument: a large-scale, deeply reconfigurable testbed, designed specifically to support cloud computing research. Chameleon differs substantially from other compute resources in that it is directly focused on supporting systems research in computer science. This type of research focuses on investigating challenges in the low level of system stack including designing new operating systems and virtualization solutions, investigating performance variability, power management, or building new networking protocols. While other NSF-sponsored scientific computing resources provide tremendous computing power for engineering and domain science research, the computers themselves and their system software components can not be changed by the end user. Similarly, resources provided by commercial clouds, introduce a virtualization layer which limits the extent to which users can modify them for use as experimental platforms. Chameleon, in contrast, offers tremendous flexibility in experiment design; users can start with "bare metal" allocations and deploy them on a range of innovative hardware types, including a variety of innovative architectures, accelerators, network equipment, and storage devices. The infrastructure is physically distributed across two main operating sites (in Chicago, Illinois and Austin, Texas) to allow researchers to replicate the multi-datacenter conditions that exist in most production science.
In addition to the hardware itself, the Chameleon project created the software, training, and user support to complete the testbed ecosystem. Unlike traditional Computer Science experimental systems which have largely been configured by technologies developed in-house, Chameleon adapted OpenStack, a mainstream open source cloud technology, to provide its capabilities. This approach has practical benefits including familiar interfaces for users and operators, workforce development, ability to leverage contributions from a large development community, and compatibility with a large ecosystem of mainstream cloud tools. At the same time, to fully support the research and sharing functions required of a scientific instrument, the Chameleon team significantly extended the capabilities of OpenStack, making substantial open source contributions to the OpenStack project, and creating new open source science-focused software distribution in the process. In phase 2, those new capabilities included in particular configurable networking, including the support for innovative Corsa switches as well as software and methodology supporting reproducibility mechanisms for computer science research. One of the most significant contributions made in this phase of the project, included the development of the initial version of CHI-in-a-Box, i.e., packaging of Chameleon Infrastructure (CHI) that ultimately allowed us to make Chameleon operations cheaper, democratize operation of complex testbeds of this type to allow other institutions to join in ? and ultimately led to integrating multiple volunteer sites, extending the pool of hardware available to our users.
To date, Chameleon has served over 8,000 users who worked on close to 1,000 unique projects in CS research, education, and innovative applications. Chameleon users came from more than 160 institutions, more than 45 countries, and their work resulted in over 600 scientific publications showcased on the Chameleon website. These publications represent advances in topics ranging from cybersecurity, scalable operating systems, power management and green computing, cloud application performance, machine learning, and many other critical areas of computing research. The Chameleon infrastructure lives on and continues to support the community through a renewal grant.
Last Modified: 06/05/2023
Modified by: Katarzyna Keahey
Please report errors in award information by writing to: awardsearch@nsf.gov.
