Research and Education:
Participants: Sam Burnett, Jason Liptak, Medha Bhadkamkar (Ph.D. student) and Raju Rangaswami (mentor)
Project Description: Efficient file systems hold one of the keys to high-performance I/O systems. Today's file systems perform a static layout of file data, aiming to preserve the directory structure of the file system and optimizing for sequential access to entire files. In this project, we re-examined the existing state-of-the-art in file system design and find it severely lacking in an important aspect, application awareness. We argued that for optimal performance, file systems must self-optimize by adapting data layout to accommodate the dynamism in application access patterns. We developed the design and implementation of an automated data layout reconfigurator which is at the heart of such a self-optimizing file system. Preliminary studies using file system traces indicated significant I/O performance gains when compared to a state-of-the-art ext3 file syst
Publications: Feasibility, Efficiency, and Effectiveness of Self-Optimizing Storage Systems Medha Bhadkamkar, Sam Burnett, Jason Liptak, Raju Rangaswami, and Vagelis Hristidis, Florida International University Technical Report TR-2007-01-01, January 2007.
Participants: Javier Ocasio Perez, Dr. S. Masoud Sadjadi (mentor)
Project Description: We define adaptability as the capacity of software in adjusting its behavior in response to changing conditions. To list just a few examples, adaptability is important in pervasive computing, where software in mobile devices need to adapt to dynamic changes in wireless networks; autonomic computing, where software in critical systems are required to be self-manageable; and grid computing, where software for long running scientific applications need to be resilient to hardware crashes and network out-ages. In this project, we investigate a realization of the transparent shaping programming model, called TRAP.NET, which enables transparent adaptation in existing .NET applications as a response to the changes in the application requirements and/or to the changes in their execution environment. Using TRAP.NET, we can adapt an application dynamically, at run time, or statically, at load time, without the need to manually modify the application original functionality-hence transparent.
Publications: S. Masoud Sadjadi and Fernando Trigoso. Trap.net: A realization of transparent shaping in .net. In Proceedings of The Nineteenth International Conference on Software Engineering and Knowledge Engineering (SEKE'2007), pages 19-24, Boston, USA, July 2007. Note: In the this paper, Javier Ocasio’s help is acknowledged.
Participants: Brittany Parsons, Ronal Stevens, Tariq King (PhD student) and Dr. Peter J. Clarke (mentor)
Project Description: This project involved the survey of several autonomic computing systems focusing on the validation methods used during self-management. The systems surveyed included Impala (Princeton University), OceanStore (UC Berkeley), Model Driven Autonomic Manager (Indiana University), and The Bison Project (University of Bologna). The results of the survey showed that there were no approaches used to validate the changes made to the systems during self-management. As a result the team decided to develop a self-testing framework for autonomic computing systems.
Participants: Joseph Marrero, Igor Hernandez, and Raju Rangaswami (mentor)
Project Description: This project developed Rootsense, a holistic and real-time intrusion prevention system that combines the merits of misbehavior-based and anomaly-based detection. Four principles govern the design and implementation of Rootsense. First, Rootsense audits events within different subsystems of the host operating system and correlates them to comprehensively capture the global system state. Second, Rootsense restricts the detection domain to root compromises only; doing so reduces run-time overhead and increases detection accuracy (root behavior is more easily modeled than user behavior). Third, Rootsense adopts a dual approach to intrusion detection -- a root penetration detector detects activities that exploit system vulnerabilities to penetrate the security perimeter, and a root misbehavior detector tracks misbehavior by root processes. Fourth, Rootsense is designed to be configurable for overhead management allowing the system administrator to tune the overhead characteristics of the intrusion prevention system that affect foreground task performance. A Linux implementation of Rootsense is analyzed for both accuracy and performance, using several real-world exploits and a range of end-host and server benchmarks.
Publications: Anatomy of a Real-time Intrusion Prevention System, Ricardo Koller, Raju Rangaswami, Joseph Marrero, Igor Hernandez, Geoffrey Smith, Mandy Barsilai, , Florida International University Technical Report TR-2007-01-01, January 2007.
Participants: Tatiana Soldo, Luis Atencio and Dr. S. Masoud Sadjadi (mentor)
Project Description: High performance computing (HPC) is gaining popularity in solving scientific applications. Using the current programming standards, however, it takes an HPC expert to efficiently take advantage of HPC facilities; a skill that a scientist does not necessarily have. This lack of separation of concerns has resulted in scientific applications with rigid code, which entangles non-functional concerns (i.e., the parallel code) into functional concerns (i.e., the core busi-ness logic). Effectively, this tangled code hinders the maintenance and evolution of these applications. In this project, we investigate the design and implementation of a software tool, called Transparent Grid Enabler (TGE), that enables separation of the task of developing the business logic of a scientific application from the task of improving its performance. TGE achieves this goal by integrating two existing software tools, namely, TRAP/J and GRID superscalar. A simple matrix multiplication program is used as a case study to demonstrate the current use and capabilities of TGE.
S. Masoud Sadjadi, J. Martinez, T. Soldo, L. Atencio, R. M. Badia, and J. Ejarque. Improving separation of concerns in the development of scientific applications. In Proceedings of The Nineteenth International Conference on Software Engineering and Knowledge Engineering (SEKE'2007), pages 456-461, Boston, USA, July 2007.
S. M. Sadjadi, J. C. Martinez, L. Atencio, T. Soldo, R. Badia, and J. Ejarque. Transparent grid enablement using transparent shaping and grid superscalar. In Poster Presented in the IBM Technology Leadership Conference, Florida International University, Miami, FL 33199, October 2006.
S. M. Sadjadi, J. C. Martinez, L. Atencio, T. Soldo, R. Badia, and J. Ejarque. Transparent grid enablement using transparent shaping and grid superscalar. In Poster Presented in the FGLSAMP Expo 2007 in Tampa, FL, February 2007. Note: This is the same poster as above, but presented in the Florida Georgia Louse Stokes Alliance for Minority Participation will be holding it's annual EXPO in February of 2007.
S. Masoud Sadjadi, Luis Atencio, and Tatiana Soldo. Trap/j v2.1: An improvement for transparent adaptation. Technical Report FIU-SCIS-2007-09-01, School of Computing and Information Sciences, Florida International University, 11200 SW 8th St., Miami, FL 33199, May 2007.
S. Masoud Sadjadi, J. Martinez, T. Soldo, L. Atencio R. M. Badia, and J. Ejarque. Improving separation of concerns in the development of scientific applications. Technical Report FIU-SCIS-2007-02-01, School of Computing and Information Sciences, Florida International University, 11200 SW 8th St., Miami, FL 33199, February 2007. Note: This technical report is an extended version of the paper published in SEKE-07.