Scalable nonblocking concurrent objects for mission critical code

Dechev, Damian; Stroustrup, Bjarne

doi:10.1145/1639950.1639954

Cited by 7 publications

(21 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such platforms could include a wide variety of features including a heterogenous design of CPUs, GPUs, and even FPGAs. In our future work we plan to implement components for supporting additional programming styles such as the partitioned global address space (PGAS) programming model [18], and nonblocking synchronization [19]. Such extensions will address the needs of applications and algorithms that increasingly rely on finegrained parallelism such as lock-free synchronization [19,20] and strong scaling while supporting fault resilience [21] to accommodate the massive growth of explicit on-chip parallelism and constrained bandwidth anticipated of future chip architectures.…”

Section: Discussionmentioning

confidence: 99%

A Macroscale Simulator for Exascale Software/Hardware Co-Design

Dechev

Hendry²

2013

J Inform Tech Soft Engg

View full text Add to dashboard Cite

The next decade will see a rapid evolution of HPC node architectures as power and cooling constraints are limiting increases in microprocessor clock speeds and constraining data movement. Future and current HPC applications will have to change and adapt as node architectures evolve. The application of advanced exascale architecture simulators will play a crucial role for the design and optimization of future data intensive applications. In this paper, we present our imulation-based framework for analyzing the scalability and performance of massive interconnected networks.

show abstract

Section: Discussionmentioning

confidence: 99%

A Macroscale Simulator for Exascale Software/Hardware Co-Design

Dechev

Hendry²

2013

J Inform Tech Soft Engg

View full text Add to dashboard Cite

show abstract

“…In particular, we have developed a collection of highly concurrent scalable data containers together with an associated compiler to support nonblocking synchronization of shared memory operations [22,47,12], a recent synchronization mechanism known to provide better safety and performance than traditional blocking synchronization techniques by eliminating hazards such as deadlock, livelock, and priority inversion and by being highly scalable in supporting large numbers of threads. The entire system integrates an extensive set of architecture-specific optimizations, e.g.…”

Section: Our Approachmentioning

confidence: 99%

“…Further, incorrect use of locks is hard to detect using traditional testing procedures, whereas a program can be deployed and used for a long period of time before the flaws become evident and cause anomalous behavior [10,12]. A key component of our research is to develop a library of composable and highly scalable generic concurrent data structures to support nonblocking synchronization [22,47,17,12], which enhances thread safety and enables fast execution by eliminating hazards associated with mutual exclusion locks.…”

Section: Designing the Scalable Template Librarymentioning

confidence: 99%

“…Lock-free and wait-free algorithms do not use mutual exclusion locks and can efficiently utilize resources on shared memory multiprocessors with low latency and high bandwidth of interprocessor communication [22]. Our existing work shows that it can eliminate whole classes of concurrency hazards while delivering significant performance improvements by a large factor for many scenarios [12,10,8,15,7,11].…”

Section: Designing the Scalable Template Librarymentioning

confidence: 99%

“…However, most of the existing concurrent programming models are based on mutual exclusion, which is the most common synchronization technique for shared data but can lead to significant overhead, high complexity, and reduced parallelism and key safety hazards including race conditions, deadlock, livelock, and ordering violations [22,12]. Such errors are hard to reproduce, often lead to unpredictable real-time behavior and are difficult to debug.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Effective software design and development for the new graph architecture HPC machines.

Dechev¹

2012

Self Cite

View full text Add to dashboard Cite

Software applications need to change and adapt as modern architectures evolve. Nowadays advancement in chip design translates to increased parallelism. Exploiting such parallelism is a major challenge in modern software engineering. Multicore processors are about to introduce a significant change in the way we design and use fundamental data structures. In this work we describe the design and programming principles of a software library of highly concurrent scalable and nonblocking data containers. In this project we have created algorithms and data structures for handling fundamental computations in massively multithreaded contexts, and we have incorporated these into a usable library with familiar look and feel.In this work we demonstrate the first design and implementation of a wait-free hash table. Our multiprocessor data structure design allows a large number of threads to concurrently insert, remove, and retrieve information. Non-blocking designs alleviate the problems traditionally associated with the use of mutual exclusion, such as bottlenecks and thread-safety. Lock-freedom provides the ability to share data without some of the drawbacks associated with locks, however, these designs remain susceptible to starvation. Furthermore, wait-freedom provides all of the benefits of lock-free synchronization with the added assurance that every thread makes progress in a finite number of steps. This implies deadlockfreedom, livelock-freedom, starvation-freedom, freedom from priority inversion, and thread-safety.The challenges of providing the desirable progress and correctness guarantees of wait-free objects makes their design and implementation difficult. There are few wait-free data structures described in the literature. Using only standard atomic operations provided by the hardware, our design is portable; therefore, it is applicable to a variety of data-intensive applications including the domains of embedded systems and supercomputers.Our experimental evaluation shows that our hash table design outperforms the most advanced locking solution, provided by Intel's TBB library, by 22%. When compared to more traditional locking designs we show a performance improvement by a factor of 7.92. When compared to alternative non-blocking designs, our hash table demonstrates solid performance gains in a large majority of cases, typically by a factor of 3.44. 3Acknowledgements

show abstract

Using SST/Macro for Effective Analysis of MPI-Based Applications: Evaluating Large-Scale Genomic Sequence Search

Dechev

Ahn

2013

IEEE Access

View full text Add to dashboard Cite

Future and current high-performance computing applications will have to change and adapt as node architectures evolve. The application of advanced architecture simulators will play a crucial role for the design and optimization of future data intensive applications. In this paper, we present our simulation-based framework for analyzing the scalability and performance of a number of critical optimizations of a massively parallel genomic search application, mpiBLAST, using an advanced macroscale simulator (SST/macro). We report the use of our framework for the evaluation of three potential improvements of mpiBLAST: 1) enabling high-performance parallel output; 2) an approach for caching database fragments in memory; and 3) a methodology for pre-distributing database segments. In our experimental setup, we performed query sequence matching on the genome of the yellow fever mosquito, Aedes aegypti.INDEX TERMS Exascale architecture simulator, mpiBLAST, performance and scalability modeling, multiprocessor programming, message passing interface.

show abstract

Scalable nonblocking concurrent objects for mission critical code

Cited by 7 publications

References 14 publications

A Macroscale Simulator for Exascale Software/Hardware Co-Design

A Macroscale Simulator for Exascale Software/Hardware Co-Design

Effective software design and development for the new graph architecture HPC machines.

Using SST/Macro for Effective Analysis of MPI-Based Applications: Evaluating Large-Scale Genomic Sequence Search

Contact Info

Product

Resources

About