VIVA -- Vertically Integrated
VirtualizAtion
Overview
Virtualization has emerged as a key system design
technology for reducing the complexity of modern hardware,
software systems, and applications, through
abstract layering that simplifies reasoning about, and making use
of, resources and system services by applications.
The benefits of virtualization
(e.g. generality, portability, extensibility, and simplifying application
debugging, deployment, and development) however, come at a cost in performance
since the number, depth, and complextity of layers is significant
and increasing.
VIVA is a comprehensive research plan that includes dynamic
compiler and runtime techniques that adaptively specialize
the software stack of a system according to the changing needs
of the application and resource performance. VIVA research
incorporates components at all layers of the system including the
virtual machine monitor, operating system, virtual execution environment,
and middleware and application server systems.
VIVA automatically integrates and optimizes together the application
and system code to extract performance levels that exceed those
currently available. VIVA is applicable to a wide range of
systems (from high-end clusters to resource-constrained devices)
and application domains (from scientific applications to business
and Internet-computing programs and services).
VIVA-Related Projects
Application-Specific Linux (ASL):
The Overhead of Paravirtualization:
Vertically Integrated Profiling
Full-System Integration and Optimization:
Download IV Heap Sizing Source and Presentation (CGO 2007)
VIVA-Related Tools
Jisha -- Automatic Guest OS Image Installation over Xen --
from within Xen
Automatic Customized Image Deployment over Xen -- from a remote admistration center
Representative VIVA-Related Publications
-
M. Weigel and C. Krintz,
Cross-Language, Type-Safe, and Transparent Object Sharing For Co-Located Managed Runtimes
To appear: OOPSLA, October, 2010
(details)
-
N. Chohan, C. Castillo, M. Spreitzer, M. Steinder, A. Tantawi, and C. Krintz,
See Spot Run: Using Spot Instances for MapReduce Workflows
To appear: HotCloud10, July, 2010
(details)
-
C. Bunch, N. Chohan, C. Krintz, J. Chohan, J. Kupferman, P. Lakhina, Y. Li, and
Y. Nomura (Fujitsu),
An Evaluation of Distributed Datastores Using the AppScale Cloud Platform,
To appear: IEEE Cloud10: International Conference on Cloud Computing, July, 2010
(details)
-
N. Chohan, C. Bunch, S. Pang, C. Krintz, N. Mostafa, S. Soman, and R. Wolski,
AppScale: Scalable and Open AppEngine Application Development and Deployment
International Conference on Cloud Computing (CloudComp), Oct, 2009,
(details)
-
Nagy Mostafa and Chandra Krintz
Tracking Performance Across Software Revisions
ACM International Conference on Principles and Practice of Programming in Java (PPPJ), Aug, 2009
(PDF)
-
Michal Wegiel and Chandra Krintz
Dynamic Prediction of Collection Yield for Managed Runtimes
ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Mar, 2009
(details)
-
Sunil Soman, Chandra Krintz, and Laurent Daynes
MTM²: Scalable Memory Management for Multi-Tasking Managed Runtime Environments
The
European Conference on Object-Oriented Programming (ECOOP), July, 2008
(details)
-
Michal Wegiel and Chandra Krintz
XMem: Type-Safe, Transparent, Shared Memory for Cross-Runtime Communication and Coordination
The ACM/SIGPLAN Conference on
Programming Language Design and Implementation (PLDI08), Jun, 2008
(PLDI), March, 2008
(details)
-
Michal Wegiel and Chandra Krintz
The Mapping Collector: Virtual Memory Support for Generational, Parallel, and Concurrent Compaction,
The International Conference on
Architectural Support for Programming Languages and Operating Systems
(ASPLOS), March, 2008
(details)
-
Chris Grzegorczyk, Sunil Soman, Rich Wolski, and Chandra Krintz,
Isla Vista Heap Sizing: Using Feedback to Avoid Paging,
The International Symposium on Code Generation and Optimization
(CGO),
March, 2007
(details)
- Hussam Mousa, Chandra Krintz, Lamia Youseff, and Rich Wolski,
VIProf: Vertically Integrated Full-System Performance Profiler
Workshop on Next-Generation Software (w/IPDPS), March, 2007
(details)
- Lamia Youseff, Rich Wolski, Brent Gorda, Chandra Krintz
Evaluating the Performance Impact of Xen on MPI and Process Execution For HPC Systems
International Workshop on Virtualization Technologies in Distributed Computing (VTDC), Nov. 2006 (w/ Supercomputing).
(PDF)
-
Selim Gurun and Chandra Krintz,
A Run-Time, Feedback-Based Energy Estimation Model For Embedded Devices,
International Conference on Hardware-Software Codesign and System Synthesis
(CODES+ISSS), October 2006, Seoul Korea
(details)
Other RACELab papers on this and related work
can be found here
.
This work is supported by:
- NSF CAREER award CNS-0546737
-- CAREER: VIVA -- Vertically Integrated VirtualizAtion: Automatic,
Full System, Specialization for High-End Computing
-
and NSF award ST-HEC-0444412 --
ST-HEC: Automatic Linux Customization and Optimization for
High-Performance Scientific Applications.
September, 2006; last update: March, 2007