Ganesh Gopalakrishnan

University of Utah

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Bruce D. Cornuelle

Scripps Institution of Oceanography

Robert M. Kirby

University of Utah

Zvonimir Rakamarić

University of Utah

Ignacio Laguna

Lawrence Livermore National Laboratory

Ibrahim Hoteit

King Abdullah University of Science and Technology

Ian Briggs

University of Utah

Quanan Zheng

University of Maryland, College Park

Sarvani Vakkalanka

University of Utah

Matthew R. Mazloff

Scripps Institution of Oceanography

Rajeev Thakur

Argonne National Laboratory

Cooperative Institutions

University of Utah

102

University of Science and Technology

Scripps Institution of Oceanography

University of California, San Diego

King Abdullah University of Science and Technology

Woods Hole Oceanographic Institution

Lawrence Livermore National Laboratory

Oceanography Society

Massachusetts Institute of Technology

University of Washington

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Decomposing the proof of correctness of pipelined microprocessors

Lecture notes in computer science (1998)

Ravi Hosabettu Mandayam Srivas Ganesh Gopalakrishnan

Abstraction

Implementation

10.1007/bfb0028739

Cite

Citations (54)

FlowFPX: Nimble Tools for Debugging Floating-Point Exceptions

arXiv (Cornell University) (2024)

Taylor Allred Xinyi Li Ashton Wiersdorf Ben Greenman Ganesh Gopalakrishnan

Reliable numerical computations are central to scientific computing, but the floating-point arithmetic that enables large-scale models is error-prone. Numeric exceptions are a common occurrence and can propagate through code, leading to flawed results. This paper presents FlowFPX, a toolkit for systematically debugging floating-point exceptions by recording their flow, coalescing exception contexts, and fuzzing in select locations. These tools help scientists discover when exceptions happen and track down their origin, smoothing the way to a reliable codebase.

10.48550/arxiv.2403.15632

Cite

Citations (0)

Systematic Debugging Methods for Large-Scale HPC Computational Frameworks

Computing in Science & Engineering (2014)

Alan Humphrey Qingyu Meng Martin Berzins Diego Caminha B. de Oliveira Zvonimir Rakamarić

Parallel computational frameworks for high-performance computing are central to the advancement of simulation-based studies in science and engineering. Unfortunately, finding and fixing bugs in these frameworks can be extremely time consuming. Left unchecked, these bugs can drastically diminish the amount of new science that can be performed. This article presents a systematic study of the Uintah Computational Framework and approaches to debug it more incisively. A key insight is to leverage the modular structure of Uintah, which lends itself to systematic debugging. In particular, the authors have developed a new approach based on coalesced stack trace graphs (CSTG) that summarize the system behavior in terms of key control flows manifested through function invocation chains. They illustrate several scenarios for how CSTGs could help efficiently localize bugs, and present a case study of how they found and fixed a real Uintah bug using CSTGs.

10.1109/mcse.2014.11

Cite

Citations (3)

Data Assimilation of Range- and Depth-Averaged Sound Speed from Acoustic Tomography Measurements in Fram Strait

Journal of Atmospheric and Oceanic Technology (2023)

Florian Geyer Ganesh Gopalakrishnan Hanne Sagen Bruce D. Cornuelle François Challet

Abstract The 2010–12 Acoustic Technology for Observing the Interior of the Arctic Ocean (ACOBAR) experiment provided acoustic tomography data along three 167–301-km-long sections in Fram Strait between Greenland and Spitsbergen. Ocean sound speed data were assimilated into a regional numerical ocean model using the Massachusetts Institute of Technology General Circulation Model–Estimating the Circulation and Climate of the Ocean four-dimensional variational (MITgcm-ECCO 4DVAR) assimilation system. The resulting state estimate matched the assimilated sound speed time series; the root-mean-squared (RMS) error of the sound speed estimate (∼0.4 m s −1 ) is smaller than the uncertainty of the measurement (∼0.8 m s −1 ). Data assimilation improved modeled range- and depth-averaged ocean temperatures at the 78°50′N oceanographic mooring section in Fram Strait. The RMS error of the state estimate (0.21°C) is comparable to the uncertainty of the interpolated mooring section (0.23°C). Lack of depth information in the assimilated ocean sound speed measurements caused an increased temperature bias in the upper ocean (0–500 m). The correlations with the mooring section were not improved as short-term variations in the mooring measurements and the ocean state estimate do not always coincide in time. This is likely due to the small-scale eddying and nonlinearity of the ocean circulation in Fram Strait. Furthermore, the horizontal resolution of the state estimate (4.5 km) is eddy permitting, rather than eddy resolving. Thus, the state estimate cannot represent the full ocean dynamics of the region. This study is the first to demonstrate the usefulness of large-scale acoustic measurements for improving ocean state estimates at high latitudes. Significance Statement Acoustic tomography measurements allow one to observe ocean temperature in large ocean volumes under the Arctic sea ice by measuring sound speed, which is hard to synoptically observe by other methods. This study has established methods for assimilation of depth- and range-averaged ocean sound speed from an acoustic tomography experiment in Fram Strait. For the first time, a 2-yr time series of ocean sound from acoustic tomography has been assimilated into an ocean state estimate. The results highlight the use of ocean tomography in ice-covered regions to improve state estimates of ocean temperature.

Mooring

Ocean observations

Eddy

Speed of Sound

10.1175/jtech-d-22-0132.1

Cite

Citations (0)

A Formal Approach to Detect Functionally Irrelevant Barriers in MPI Programs

Lecture notes in computer science (2008)

Subodh Sharma Sarvani Vakkalanka Ganesh Gopalakrishnan Robert M. Kirby Rajeev Thakur

Message Passing Interface

Interleaving

Statement (logic)

Problem statement

10.1007/978-3-540-87475-1_36

Cite

Citations (19)

A Scalable and Distributed Dynamic Formal Verifier for MPI Programs

Anh Vo Sriram Aananthakrishnan Ganesh Gopalakrishnan Bronis R. de Supinski Martin Schulz

Standard testing methods of MPI programs do not guarantee coverage of all non-deterministic interactions (e.g., wildcard-receives). Programs tested by these methods can have untested paths (bugs) that may become manifest unexpectedly. Previous formal dynamic verifiers cover the space of non-determinism but do not scale, even for small applications. We present DAMPI, the first dynamic analyzer for MPI programs that guarantees scalable coverage of the space of non-determinism through a decentralized algorithm based on Lamport-clocks. DAMPI computes alternative non-deterministic matches and enforces them in subsequent program replays. To avoid interleaving explosion, DAMPI employs heuristics to focus coverage to regions of interest. We show that DAMPI can detect deadlocks and resource-leaks in real applications. Our results on a wide range of applications using over a thousand processes, which is an order of magnitude larger than any previously reported results for MPI dynamic verification tools, demonstrate that DAMPI provides scalable, user-configurable testing coverage.

Interleaving

Heuristics

10.1109/sc.2010.7

Cite

Citations (93)

Toward Increasing Trust in Exascale Simulations

Dorra Ben Khalifa Xinyi Li Ignacio Laguna Matthieu Martel Ganesh Gopalakrishnan

In recent decades, High Performance Computing (HPC) and simulations have become determinant in many areas of engineering and science. Since many HPC applications rely extensively on floating-point arithmetic operations to solve computational problems, many kinds of numerical errors can be introduced during the program execution, leading to instability or reproducibility problems. One kind of these error sources is the loss of significant digits or cancellation which produces inaccurate results when two nearby numbers are subtracted. In this article, we present Candy, a new dynamic library based on code instrumentation that detects cancellations in numerical software. The originality of our method is to compute the number of significant bits of floating-point numbers in a generalized framework by attaching a shadow value in higher precision to each number. This helps to detect in an accurate way if a program suffers from cancellation problems and thus to increase the trust in large-scale HPC applications and exascale simulations. We evaluate Candy over a set of complex and real-world numerical applications. In addition, we compare our method against the state-of-art tool FPChecker in terms of efficiency, mixed precision results and speed of the analysis.

Code (set theory)

10.1109/xloop56614.2022.00010

Cite

Citations (0)

Sparse Matrix Code Dependence Analysis Simplification at Compile Time

arXiv (Cornell University) (2018)

Mahdi Soltan Mohammadi Kazem Cheshmi Ganesh Gopalakrishnan Mary Hall Maryam Mehri Dehnavi

Analyzing array-based computations to determine data dependences is useful for many applications including automatic parallelization, race detection, computation and communication overlap, verification, and shape analysis. For sparse matrix codes, array data dependence analysis is made more difficult by the use of index arrays that make it possible to store only the nonzero entries of the matrix (e.g., in A[B[i]], B is an index array). Here, dependence analysis is often stymied by such indirect array accesses due to the values of the index array not being available at compile time. Consequently, many dependences cannot be proven unsatisfiable or determined until runtime. Nonetheless, index arrays in sparse matrix codes often have properties such as monotonicity of index array elements that can be exploited to reduce the amount of runtime analysis needed. In this paper, we contribute a formulation of array data dependence analysis that includes encoding index array properties as universally quantified constraints. This makes it possible to leverage existing SMT solvers to determine whether such dependences are unsatisfiable and significantly reduces the number of dependences that require runtime analysis in a set of eight sparse matrix kernels. Another contribution is an algorithm for simplifying the remaining satisfiable data dependences by discovering equalities and/or subset relationships. These simplifications are essential to make a runtime-inspection-based approach feasible.

Leverage (statistics)

Compile time

Dependence analysis

Matrix (chemical analysis)

Code (set theory)

10.48550/arxiv.1807.10852

Cite

Citations (0)

Session details: Session order 5: synchronization session

Ganesh Gopalakrishnan

No abstract available.

10.1145/3250902