Systems and Architecture

 

All aspects of computer systems have seen rapid advances in recent years and are still evolving. This makes major impacts on diverse demanding applications of computers spanning from micro-scale engineering applications to large-scale scientific applications in industry and academic communities. The systems and architecture group encompasses the structure, organization, implementation, and performance evaluation of the different building blocks of computer and network systems. Our network systems research covers a broad range of topics including wireless sensor networks, high-speed networks, network security, routing, mobile systems, and network simulations. Also our research interests are in distributed systems, multiprocessor systems, embedded and robotics systems, and relevant operating systems and security.

 

Faculty


Gerald Baumgartner: Embedded Systems Programming, Distributed Systems

Costas Busch: Multiprocessor Systems, Transactional Memory

Feng Chen: Operating Systems, Storage Systems

Rajgopal Kannan: Wireless Sensor Networks, Network Security

Supratik Mukhopadhyay: Robotics

Seung-Jong Park: High Speed Networks, Wireless Sensor Networks

Brygg Ullmer: Tangible Devices, Embedded Systems

Hartmut Kaiser (adjunct): Future systems, Exascale, Parallelism and Concurrency in C++

Clint Whaley: Floating Point Optimization

Qingyang Wang: Distributed Systems, Cloud Computing

 

Specific Projects


Virtual embedded systems testbed. Baumgartner is developing a hardware-in-the-loop simulation environment for embedded systems for use both as a teaching tool and as a software development and testing tool. This simulation environment allows programming an actual embedded processor, while some or all of the electrical and mechanical devices controlled by the embedded processor, such as power switches or motors, are simulated or monitored on the PC.

Organic grid. Baumgartner investigates a biologically inspired and fully-decentralized approach to the organization of computation that is based on the autonomous scheduling of strongly mobile agents on a peer-to-peer network. Every node in the network is equally responsible for scheduling and computation.

Transactional memory systems. Busch studies a new distributed programming model where sequences of shared memory operations are wrapped in atomic transactions similar to classic data base transactions. He provides novel efficient schedulers for transactional memory systems both in tightly coupled architectures and in distributed systems.

Persistent memory/storage systems. F Chen studies the impact of emerging persistent memory/storage technologies (e.g., NAND flash, PCM) to system architecture designs, from perspectives of performance, reliability, and power efficiency. The research also includes developing system mechanisms for effectively integrating PM in the memory and storage hierarchy. Supported by Louisiana BoR and NSF.

Automated docking of ships. Sponsored by Marinetronix Ltd.

Secure infrastructure for networked systems. Supported by LSU-EDA program.

Cognitive radios. Kannan works on developing optimal resource allocation policies for cognitive radios that coexist/compete for the same shared spectrum using techniques from game-theory and online algorithms. Funded by AFOSR.

High-speed networks. Park's CRON project investigates 10Gbps high-speed networks (http://www.cron.loni.org). His GENI (Global Environment for Network Innovation) project develops network protocols (http://www.geni.net). These projects are supported by NSF.

Wireless networks. Park develops energy-efficient survivable and secure routing protocols for distributed sensor networks; and collaborates on interoperable cognitive radio platform for data transmission in military related applications. Supported by ONR.

Exascale Computing. Current software solutions are unable to allow users to fully utilize the system's hardware potential. Only by embracing new programming models will applications be able to utilize the potential of exascale machines. Hartmut Kaiser along with the STE||AR Group (http://stellar.cct.lsu.edu/) is leading the development of HPX, a C++ runtime system which challenges the paradigms of today to scale on the machines of tomorrow.

Floating Point Optimization. Whaley does research on backend compilation with the iFKO (iterative Floating point Kernel Optimizer) compilation framework, which is released as part of the ATLAS project (see Computational Science).

Large-scale web applications in Cloud. Q Wang studies the performance and scalability issues of large-scale web applications (e.g., Amazon) running in the cloud. The essential goal of this project is to build the next generation web applications that can simultaneously achieve good performance and high resource efficiency in cloud computing platforms. His research is Supported by Louisiana BoR and NSF.

Application-layer DDoS attacks. Q Wang investigates new types of application layer DDoS attacks for internet services. He is developing mechanisms of launching low-volume application layer DDoS attacks which can cause significant performance damage (e.g., the long-tail latency problem) of the target internet service while still keep stealthy. His research is Supported by Louisiana BoR and NSF.