| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | March 1, 2010 |
| Latest Amendment Date: | June 30, 2014 |
| Award Number: | 0953541 |
| Award Instrument: | Continuing Grant |
| Program Manager: |
Marilyn McClure
mmcclure@nsf.gov (703)292-5197 CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | August 1, 2010 |
| End Date: | July 31, 2016 (Estimated) |
| Total Intended Award Amount: | $418,961.00 |
| Total Awarded Amount to Date: | $418,961.00 |
| Funds Obligated to Date: |
FY 2011 = $80,131.00 FY 2012 = $83,621.00 FY 2013 = $87,280.00 FY 2014 = $91,124.00 |
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
201 OLD MAIN UNIVERSITY PARK PA US 16802-1503 (814)865-1372 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
201 OLD MAIN UNIVERSITY PARK PA US 16802-1503 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | CSR-Computer Systems Research |
| Primary Program Source: |
01001112DB NSF RESEARCH & RELATED ACTIVIT 01001213DB NSF RESEARCH & RELATED ACTIVIT 01001314DB NSF RESEARCH & RELATED ACTIVIT 01001415DB 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
The cloud computing model has opened up new possibilities for the realization of the long-cherished goal of utility computing. Utility computing represents the desire to have IT acquired, delivered, used, paid for, and managed in a manner similar to the way we use other commoditized utilities. The principal appeal of utility computing lies in the systematized framework it could create for the interaction between providers and consumers of IT resources. In particular, utility computing should enable consumers to participate in active and informed ways in making resource procurement decisions in a transparent ``market'' of competing providers. Consumers of current cloud-based offerings have a limited view of and control over resource procurement and control, a significant hindrance in the realization of a utility. This research will develop mechanisms and techniques that would reduce this gap, thereby helping turn cloud-based offerings of the near future into mature utilities.
This research will define, formulate, and solve fundamentally novel resource management problems---consumer-end metering, auditing, and dynamic mapping between virtual and physical resources---in cloud computing. It will result in novel utility-enabling facilities that will reduce the burden on application developers and system administrators wishing to outsource their IT needs to the cloud by easing and automating currently non-existent or inadequate decision-making related to resource procurement and modulation. Prototypes and source code will be shared with other researchers for independent use, experimentation, and deployment. The plan for integration of teaching and research will consist of (i) engaging undergraduates in research through REU supplements, (ii) design of new graduate courses on this topic with projects based on cloud-hosted teaching testbeds, and (iii) dissemination among researchers and industry via open-source software and workshops.
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.
Summary of outcomes: This project developed systems and algorithmic support to help convert recently emergent public cloud computing platforms into enablers of utility computing. Towards this, the PI and his students studied problems concerning the design and operation of both key actors in such a computing environment - the public cloud provider and its tenant customer workload. All the source code and data from this work was made publicly available and published in reputed systems and performane evaluation conferences and journals. A combination of theoretical analysis and prototype design was used to evaluate the efficacy of proposed ideas in the context of real-world application workloads and cloud providers.
Intelletual merits: This project made novel contributions towards establishing the feasibility of public cloud platforms as enablers of utility computing. Key contributions on the cloud provider side were: (i) facilities for accounting resource usage on a per-tenant basis in consolidated settings, (ii) algorithms for improving the operational efficiency of the cloud provider applied both to the IT infrastructure of its data centers and its power delivery infrastructure, (iii) the use of dynamic pricing and dynamic resource allocation (the latter labeled "effective capacity" modulation) as knobs for profit optimization. Key contributions on the tenant side were: (i) the design of modeling and optimization techniques for a variety of tenants tooperate cost-effectively on highly dynamic public cloud offerings, (ii) evaluation of the efficacy of these ideas in the specific context of several concrete workload types (distributed in-memory caching basedon memcached, web servers, and parallel big data processing based on Apache Spark) leveraging cheap spot and burstable instances for cost-efficacy. All proposed ideas were prototyped and experiments were carried out on real-world public cloud platforms (Amazon EC2 and Google Compute Engine) to demonstrate their efficacy.
Broader impacts: The PI developed a series of 3 graduate-level seminar courses on cloud computing (Fall 2010 on general overview, Spring 2014 focusing on performance evaluation, and Spring 2017 focusing on public cloud computing). These will form the basis of a new core graduate course that will be offered in the PI's department on a regular basis. Concepts related to cloud computing were also included in both the undergraduate and graduate operating systems courses that PI regularly teaches. The project supported one Ph.D. student's research and two MS students (one woman among them). Finally, the PI engaged in outreach via (i) collaborations with industrial collaborators at IBM Zurich, IBM Almaden, and Microsoft Research and (ii) invited talks at Microsoft Faculty Summit 2014, Cisco NAG 2012, Lund University Cloud Control Conference 2014, and a keynote talk at Jawaharlal Nehru University (New Delhi India) in December 2015.
Last Modified: 12/29/2016
Modified by: Bhuvan Urgaonkar
Please report errors in award information by writing to: awardsearch@nsf.gov.
