News People Recent Publications Projects

Welcome to SSS@Lehigh!
We are the Scalable Systems and Software Research Group at Lehigh University!
Designing and building parallel, concurrent, and distributed systems is our major focus. Our research topics include:


Programming Languages
Transactional Memory
Concurrent Data Strutures
Speculative Parallelization
Memory Allocators
LLVM
Memory Persistence
Distributed Synchronization
Centralized and Distributed Database
Multicore and NUMA Synchronization
Distributed Transactions
Transactional Systems
Consistency
Fault-Tolerance and Replication
Geographic Replication
Consensus
Computer Architecture
Energy Efficient Architetures
We are always looking for motivated students and productive collaborations. If you find these topics interesting and you are eager to contribute, contact Prof. Korth or Prof. Spear or Prof. Palmieri or Prof. Guo.

Latest News

  • Our paper "HaTS: Hardware-assisted Transaction Scheduler" has been accepted at OPODIS 2019. Congratulations Zhanhao!
  • Our paper "Optimizing Persistent Memory Transactions" has been accepted at PACT 2019. Congratulations Pantea and Tingzhe!
  • Prof. Palmieri will serve as committee member of Usenix ATC 2020. Consider to submit and attend!
  • Our paper "Scrabble: A Fine-¬≠Grained Cache with Adaptive Merged Block" has been accepted at IEEE Transactions on Computers. Congratulations Chao and Yuan!
  • Prof. Korth and Prof. Palmieri have been invited to present at the 2019 Lehigh Quant/Financial Engineering Conference.
  • Prof. Korth serves as committee member of BCDL 2020. Consider to attend!
  • Prof. Guo will serve as committee member of IPDPS 2020. Consider to submit and attend!
  • Prof. Spear will serve as Chair of the Program Committee of SPAA 2020. Consider to submit and attend!
  • Prof. Palmieri will serve as Local Arrangement Chair of SPAA 2020 to be held in Philadelphia, PA, USA.
  • Our paper "On the Correctness of Transaction Processing with External Dependency" has been accepted as Brief Announcement at DISC 2019. Congrats Masoomeh!
  • Prof. Guo will serve as committee member of HPCA 2020. Consider to submit and attend!
  • Prof. Palmieri will serve as committee member of ICPP 2020. Consider to submit and attend!
  • Our paper "A Practical, Scalable, Relaxed Priority Queue" has been accepted at ICPP 2019. Congratulations Tingzhe!
  • Prof. Spear will serve as committee member of PPoPP 2020. Consider to submit and attend!
  • Our paper "Simplifying Transactional Memory Support in C++" has been accepted at ACM TACO. Congratulations Pantea!
  • Our PhD student Tingzhe Zhou successfully defended his PhD disseretation. Tingzhe will join Facebook. Congratulations Tingzhe and good luck!
  • Our paper "SSS: Scalable Key-Value Store with External Consistent and Abort-free Read-only Transactions" has been accepted at ICDCS 2019. Congratulations Masoomeh!
  • Our paper "Optimizing Persistent Transactions" has been accepted as Brief Announcement at SPAA 2019. Congratulations Tingzhe and Pantea!
  • Prof. Spear will serve as committee member of DISC 2019. Consider to submit and attend!
  • Our paper "Lerna: Parallelizing Dependent Loops Using Speculation" has been accepted for publication at ACM Transactions on Storage, ACM ToS, 2019.
  • Prof. Spear will serve as committee member of PACT 2019. Consider to submit and attend!
  • Our MS student Yaodong Sheng is now a PhD student at SSS. Welcome Yaodong!
  • Our paper "Don't Forget About Synchronization! A Case Study of K-Means on GPU" has been accepted at PMAM 2019. Congratulations Jacob!
  • The abstract submitted by our PhD student Jacob to the CGO Student Research Competition on the performance interaction betweeb RDMA and NUMA has been accepted. Congratulations Jacob!
  • Prof. Palmieri will serve as committee member of ACM PaPoC 2019. Consider to submit and attend!
  • Our paper "Processing Transactions in a Predefined Order" has been accepted at PPoPP 2019.
  • Our PhD students Masoomeh and Pantea attended the 2018 Grace Hopper Celebration (GHC), the world's largest gathering of women technologists.
  • Our paper "NUMASK: High Performance Scalable Skip List for NUMA" has been accepted at DISC 2018.
  • Our NSF project "Data Structure Designs and Implementations for Modern Multicore Architectures and Applications" has been funded for three years!
  • Prof. Palmieri will serve as committee member of ACM SYSTOR 2019. Consider to submit and attend!
  • Prof. Palmieri will serve as external committee member of ACM SIGPLAN PPoPP 2019. Consider to submit and attend!
  • Prof. Spear will serve as committee member of ACM SIGPLAN PPoPP 2019. Consider to submit and attend!
  • Our PhD student Tingzhe Zhou passed his PhD general exam. Congratulations Tingzhe!

People

Faculty:

Hank Korth

Hank Korth

Professor
Michael Spear

Michael Spear

Associate Professor
Roberto Palmieri

Roberto Palmieri

Assistant Professor
Xiaochen Guo

Xiaochen Guo

Assistant Professor

Postdoc:

Ahmed Hassan

Ahmed Hassan

Postdoctoral Resedarch

Current Students:

Past Students:

  • Tingzhe Zhou - PhD 2019 - First employment: Facebook
  • Zhanhao Chen - MS 2018 - First employment: Palo Alto Networks
  • Shihao Jing - MS 2018 - First employment: Google
  • Michael Embick - MS 2018 - First employment: Lockheed Martin
  • Henry Daly - Undergraduate 2018 - First employment: Lockheed Martin
  • Chao Wang - MS 2016 - First Employment: USC
  • Wenjia Ruan - PhD 2015 - First employment: Qualcomm Research
  • Yujie Liu - PhD 2015 - First employment: Google
  • Steve Louie - MS 2015 - First employment: Cisco
  • Adam Schaub - MS 2015 - First employment: Cisco

Recent Publications

  • Zhanhao Chen, Ahmed Hassan, Masoomeh Javidi Kishi, Jacob Nelson, Roberto Palmieri
    HaTS: Hardware-assisted Transaction Scheduler
    The 23st International Conference on Principles of Distributed Systems, OPODIS, December 17-19, 2019, Neuchatel, Switzerland.
  • Chao Zhang, Yuan Zeng, Xiaochen Guo
    Scrabble: A Fine-­Grained Cache with Adaptive Merged Block
    IEEE Transactions on Computers, IEEE TC.
  • Pantea Zardoshti, Tingzhe Zhou, Yujie Liu, Michael Spear
    Optimizing Persistent Memory Transactions
    The 28th International Conference on Parallel Architectures and Compilation Techniques, PACT, September 23 - 26, 2019, Seattle, WA, USA.
  • Henry F. Korth
    Consensus in Enterprise and Financial Blockchains: Assumptions and Challenges
    The 1st Workshop on Blockchain and Distributed Ledger, BCDL, August 30, 2019, Los Angeles, California, USA.
  • Hesam Shabani and Xiaochen Guo
    ClusCross: A New Topology for Silicon Interposer-Based Network-on-Chip
    The 13th IEEE/ACM International Symposium on Networks-on-Chip, NOCS, October 17-18, 2019, New York, USA.
  • Masoomeh Javidi Kishi, Ahmed Hassan, Roberto Palmieri
    Brief Announcement: On the Correctness of Transaction Processing with External Dependency
    The 33rd International Symposium on Distributed Computing, DISC, October 14-18, 2019, Budapest, Hungary.
  • Tingzhe Zhou, Maged Michael, and Michael Spear
    A Practical, Scalable, Relaxed Priority Queue
    The 48th International Conference on Parallel Processing, ICPP, August 5-8, 2019, Kyoto Research Park, Kyoto, Japan.
  • Pantea Zardoshti, Tingzhe Zhou, Pavithra Balaji, Michel L. Scott, Michael Spear
    Simplifying Transactional Memory Support in C++
    ACM Transactions on Architecture and Code Optimization, TACO.
  • Masoomeh Javidi Kishi, Sebastiano Peluso, Hank Korth, Roberto Palmieri
    SSS: Scalable Key-Value Store with External Consistent and Abort-free Read-only Transactions
    The 39th IEEE International Conference on Distributed Computing Systems, ICDCS, July 7-10, 2019, Dallas, TX, USA.
  • Tingzhe Zhou, Pantea Zardoshti, Michael Spear
    Brief Announcement: Optimizing Persistent Transactions
    The 31st ACM Symposium on Parallelism in Algorithms and Architectures, SPAA, June 22-24, 2019, Phoenix, AZ, USA.
  • Jacob Nelson, Roberto Palmieri
    Understanding RDMA Behavior in NUMA Systems
    The 2019 Edition of the International Symposium on Code Generation and Optimization, CGO, Student Research Competition, 2019.
  • Mohamed Saad, Roberto Palmieri, Binoy Ravindran
    Lerna: Parallelizing Dependent Loops Using Speculation
    ACM Transactions on Storage, ACM ToS, 2019.
  • Jacob Nelson, Roberto Palmieri
    Don't Forget About Synchronization! A Case Study of K-Means on GPU
    The 10th International Workshop on Programming Models and Applications for Multicores and Manycores, PMAM, February 17, 2019, Washington DC, USA.
  • Mohamed M. Saad, Masoomeh Javidi Kishi, Shihao Jing, Sandeep Hans, Roberto Palmieri
    Processing Transactions in a Predefined Order
    The 24th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP, February 16-20, 2019, Washington DC, USA.
  • Henry Daly, Ahmed Hassan, Michael Spear, Roberto Palmieri
    NUMASK: High Performance Scalable Skip List for NUMA
    The 32nd International Symposium on Distributed Computing, DISC, October 15-19, 2018, New Orleans, Louisiana, USA.
  • Pantea Zardoshti, Michael Spear
    Universal Support for Scoped Memory Access Instrumentation
    The 2019 Edition of the Grace Hopper Celebration, GHC, September 26-28, 2018, Houston, Texas, USA.
  • Mohamed Mohamedin, Sebastiano Peluso, Masoomeh Javidi Kishi, Ahmed Hassan, Roberto Palmieri
    Nemo: NUMA-aware Concurrency Control for Scalable Transactional Memory
    The 47th International Conference on Parallel Processing, ICPP, August 13-16, 2018, Eugene, Oredon, USA.
  • Pantea Zardoshti, Michael Spear
    Lightweight Language-Level Support for Transactional Memory
    The 39th ACM SIGPLAN Conference on Programming Language Design and Implementation, PLDI, June 18 - 22, 2018, Philadelphia, PA, USA.
  • Mohamed Saad, Roberto Palmieri, Binoy Ravindran
    Lerna: Parallelizing Dependent Loops Using Speculation
    The 11th ACM International Systems and Storage Conference, SYSTOR, June 4-6, 2018, Haifa, Israel. Best Paper AwardBest Paper AwardBest Paper Award
  • Tingzhe Zhou, Victor Luchangco, Michael Spear
    Extending Transactional Memory with Atomic Deferral
    The 21st International Conference on Principles of Distributed Systems, OPODIS, December 18 - 20, 2017, Liston, Portugal. 
  • Mohamed Mohamedin, Masoomeh Javidi Kishi , Roberto Palmieri
    Shield: a Middleware to Tolerate CPU Transient Faults in Multicore Architectures
    16th IEEE International Symposium on Network Computing and Applications, NCA, October 30 - November 1, 2017, Boston, USA.
  • Louis Jenkins, Tingzhe Zhou, Michael Spear
    Redesigning Go's Built-In Map to Support Concurrent Operations
    The 26th International Conference on Parallel Architectures and Compilation Techniques, PACT, September 9 - 13, 2017, Portland, Oregon, USA.
  • Tingzhe Zhou, Pantea Zardoshti, Michael Spear
    Practical Experience with Transactional Lock Elision
    The 46th International Conference on Parallel Processing, ICPP, August 14 - 17, 2017, Bristol, UK.
  • Tingzhe Zhou, Victor Luchangco, Michael Spear
    Brief Announcement: Extending Transactional Memory with Atomic Deferral
    The 29th ACM Symposium on Parallelism in Algorithms and Architectures, SPAA, July 24 - 26, 2017, Washington D.C., USA.
  • Tingzhe Zhou, Victor Luchangco, Michael Spear
    Hand-Over-Hand Transactions with Precise Memory Reclamation
    The 29th ACM Symposium on Parallelism in Algorithms and Architectures, SPAA, July 24 - 26, 2017, Washington D.C., USA.
  • Balaji Arun, Sebastiano Peluso, Roberto Palmieri, Giuliano Losa, Binoy Ravindran
    Speeding up Consensus by Chasing Fast Decisions
    47th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN, June 26-29, 2017, Denver, CO, USA.
  • Xiaochen Guo, Aviral Shrivastava, Michael Spear, Gang Tan
    Languages Must Expose Memory Heterogeneity
    The 2nd International Symposium on Memory Systems, MemSys, October 4 - 6, 2016, Alexandria, Virginia, USA.
  • Sandeep Hans, Ahmed Hassan, Roberto Palmieri, Sebastiano Peluso, Binoy Ravindran
    Opacity vs TMS2: Expectations and Reality
    30th International Symposium on Distributed Computing, DISC, September 26 - 30, 2016, Paris, France.
  • Adam Schaub, Michael Spear
    Towards Migrating Computation to Distributed Memory Caches
    The International Workshop on Distributed Cloud Computing, DCC, co-located with PODC 2016, July 25, 2016, Chicago, Illinous, USA.
  • Sebastiano Peluso, Alexandru Turcu, Roberto Palmieri, Giuliano Losa, Binoy Ravindran
    Making Fast Consensus Generally Faster
    46th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN, June 28 - July 1, 2016, Toulouse, France.
  • Duane Niles, Roberto Palmieri, Binoy Ravindran
    Exploiting Parallelism of Distributed Nested Transactions
    The 9th ACM International Systems and Storage Conference, SYSTOR, June 6-8, 2016, Haifa, Israel.
  • Mohamed Saad, Roberto Palmieri, Ahmed Hassan, Binoy Ravindran
    Extending TM Primitives using Low Level Semantics
    The 28th ACM Symposium on Parallelism in Algorithms and Architectures, SPAA, July 11 - 13, 2016, Asilomar State Beach, California, USA.
  • Chao Wang, Michael Spear
    Practical Condition Synchronization for Transactional Memory
    The 11th European Conference on Computer Systems, EuroSys, April 18 - 21, 2016, London, UK
  • Tingzhe Zhou, Michael Spear
    The Mimir Approach to Transactional Output
    The 11th ACM SIGPLAN Workshop on Transactional Computing, TRANSACT, March 12, 2016, Barcelona, Spain
  • Mohamed Saad, Roberto Palmieri, Binoy Ravindran
    Lerna: Transparent and Effective Speculative Loop Parallelization
    The 11th ACM SIGPLAN Workshop on Transactional Computing, TRANSACT, March 12, 2016, Barcelona, Spain
  • Mohamed Saad, Roberto Palmieri, Ahmed Hassan, Binoy Ravindran
    On Extending TM Primitives using Low Level Semantics
    The 11th ACM SIGPLAN Workshop on Transactional Computing, TRANSACT, March 12, 2016, Barcelona, Spain
  • Mohamed Mohamedin, Roberto Palmieri, Sebastiano Peluso, Binoy Ravindran
    On Designing NUMA-Aware Concurrency Control for Scalable Transactional Memory
    The 21st ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP, Poster paper, March 12-16, 2016, Barcelona, Spain.
  • Mohamed Saad, Roberto Palmieri, Binoy Ravindran
    On Ordering Transaction Commit
    The 21st ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP, Poster paper, March 12-16, 2016, Barcelona, Spain.

(Some of our) Projects

Transactional Memory

A Full-Stack Ecosystem for Transactional Memory

We are exploring all aspects of support for transactional programming, to include compiler support, run-time mechanisms for transactional I/O, transactional workloads, and language integration.
Github link

Lightweight Abstract Memory Features

Lightweight Abstract Memory Features

We have designed an abstract instrumented memory interfaces that allow programmers to exploit hardware and software features of transactional memory as well as persistent memory. We used the LLVM compiler framework for program analysis and transformation.
Github link

NUMA-aware Data Structure

NUMA-aware Data Structures for Modern Multicore Architectures

In this project we explore design and implementation of concurrent data structures optimized to provide high performance on large multicore architectures (> 100 cores), where the memory latency distribution is not uniform, and distributed deployments.

Snow

Semantic Extension for Concurrent Applications

Concurrent Applications include blocks of code ensuring threads mutual exclusion to preserve correct accesses to data. In multicore architectures, the traditional way to implement mutual exclusion leads to resource underutilization. In this project we exploit application semantics to improve performance of atomic parts of code using compiler and library extensions.

Snow

Distributed Databases

Distributed deployments (e.g., Cloud Computing) are the only feasible way to meet elasticity and scalability requirements of large applications and systems. We focus on designing scalable transactional databases that guarantee strong correctness high-level properties, such as non-blocking queries, to retain high performance.


Support

This work is supported in part by US National Science Foundation under grants CNS-1814974, and AFOSR under grants FA9550-17-1-0367. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or the United States Air Force.