About

We are interested in the design of systems software for high-performance and parallel computing. As we see hardware scale up--both in terms of the number of interconnected machines and in terms of the number of computing elements on a single chip--We believe it is time to reconsider some of the design decisions that have sustained the popularity of the Linux/x86 software stack for general-purpose computing. Particularly, how should we organize/structure systems software stacks for future massive-scale machines?

Follow @HExSA-Lab

Alumni

Undergraduates

• Justin Goodman (Summer '19 REU student, project: Process state space dynamics)
• Justin Orr (Summer '19, project: Bringing the hybrid virtual machine model to KVM)
• Gyucheon Heo (Summer '19, project: Exploration of filesystem abstraction overheads in Linux)
• Lucas Myers (Summer '18, project: NES extensions to Hawknest full system emulator)
• Sam Grayson (Summer '18 REU student, project: VM-based query execution engines paired with specialized OSes)
• Jagruti Depan (Summer '18 REU student, project: Exploration of network science algorithms on FPGAs)
• Zachary McKee (now at Northwestern University)
• Josue Rodriguez Nieves (Summer '17 REU student - project: Reconfigurable Network-on-Chip Architectures)

Masters Students

• Piyush Nath (M.S. 2018) (Project: Prototype InfiniBand driver for Nautilus)
• Ganesh Mahesh (M.S. 2018) (Project: Address space dynamics)
• Goutham Kannan (M.S., 2017, now at Sling TV) (Project: Lua support in Nautilus)
• Imran-Ali Usmani (M.S., 2017, now Software Engineer at Argonne National Labs) (Project: Lua support in Nautilus)
• Suraj Chafle (M.S., 2017, now at Groupon) (Project: Dune port to Nautilus)

Publications

Selected Publications

Refereed Conference and Workshop Publications

Non-Overlapping Technical Reports

Projects

Hybrid Runtimes

Hybrid Runtimes are a combination of a parallel runtime and a traditional OS kernel layer called an Aerokernel, which is similar to a library OS, but fully implemented in kernel-mode. This mash-up of the runtime and OS kernel allows the runtime to choose its own abstractions and access privileged hardware which is typically only accessed at ring 0. Hybrid runtimes are a good match for high-performance systems where performance is critical. Take a look at the case paper here.

Nautilus Aerokernel

Nautilus is an Aerokernel that we built to support the Legion runtime and other home-grown parallel runtimes. It is designed for the hybrid runtime model and supports large many-core chips with complex memory hierarchies. It boots on commodity x86-64 hardware and the Intel Xeon Phi. You can find out more at the Nautilus website.

Multiverse Runtime System

Multiverse is a runtime system that enables legacy applications to run in the context of a Hybrid Runtime (HRT) environment without any porting effort. The runtime (with the application) is automatically split into two parts. One part provides a legacy general-purpose OS and library environment. The other part constitutes an accelerated environment wherein the runtime enjoys direct hardware access and the light-weight, high-performance mechanisms provided by an Aerokernel.

Nemo Event System

Nemo is an accelerated, asynchronous software event system designed for the Hybrid Runtime (HRT) model. Nemo events leverage Inter-processor Interrupt (IPI) hardware and the privileged execution model of HRTs to speed up software events by several orders of magnitude.

Sponsors

Our primary funding currently comes from the National Science Foundation via the following awards:

• NSF CNS-1763612: CSR: Medium: Collaborative Research: Interweaving the Parallel Software/Hardware Stack. Collaborative proposal with Peter Dinda, Nikos Hardavellas, and Simone Campanoni at Northwestern University.
• NSF CNS-1718252: CSR: Small: Collaborative Research: Flexible Resource Management and Coordination Schemes for Lightweight, Rapidly Deployable OS/Rs. Collaborative proposal with Jack Lange at the University of Pittsburgh.
• NSF CNS-1730689: CRI: II-NEW: MYSTIC: Programmable Systems Research Testbed to Explore a Stack-WIde Adaptive System fabriC.
• NSF CCF-1757964: REU Site: Collaborative Research: BigDataX: From theory to practice in Big Data computing at eXtreme scales. Collaborative REU with UChicago.

Some ongoing FPGA work is enabled through access to the HARPv2 (now vLab) testbed, courtesy of Intel.