Venkatramani Balaji
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Venkatramani Balaji has a total of 209 co-authors. Here are the top co-authors by number of shared publications:
- Bryan N. Lawrence
National Centre for Atmospheric Science, Department of Meteorology, University of Reading, Reading, UK
12 shared publications - Redouane Lguensat
Institut Pierre-Simon Laplace
8 shared publications - Eric Guilyardi
LOCEAN‐IPSL Sorbonne Université CNRS IRD MNHN Paris France
7 shared publications - Julie Deshayes
Sorbonne Université-CNRS-IRD-MNHN, LOCEAN Laboratory, Paris, France
5 shared publications - Karl E. Taylor
PCMDI, Lawrence Livermore National Laboratory, Livermore, CA, USA
5 shared publications Arnaud Caubel
Institut Pierre-Simon Laplace, 11 Bd d'Alembert, 78280 Guyancourt, France
4 shared publications
Latest Publications
- Open Access
- 26 December 2024
- Published by The Open Journal in Journal of Open Source Education
- Vol. 7 (82) , 241
Learning Machine Learning with Lorenz-96
- 26 December 2024
- Published by The Open Journal in Journal of Open Source Education
- Vol. 7 (82) , 241
Open AccessLearning Machine Learning with Lorenz-96
- Open Access
- 19 April 2024
- Published by Copernicus GmbH in Geoscientific Model Development
- Vol. 17 (8) , 3081-3098
The computational and energy cost of simulation and storage for climate science: lessons from CMIP6
The Coupled Model Intercomparison Project (CMIP) is one of the biggest international efforts aimed at better understanding the past, present, and future of climate changes in a multi-model context. A total of 21 model intercomparison projects (MIPs) were endorsed in its sixth phase (CMIP6), which included 190 different experiments that were used to simulate 40 000 years and produced around 40 PB of data in total. This paper presents the main findings obtained from the CPMIP (the Computational Performance Model Intercomparison Project), a collection of a common set of metrics, specifically designed for assessing climate model performance. These metrics were exclusively collected from the production runs of experiments used in CMIP6 and primarily from institutions within the IS-ENES3 consortium. The document presents the full set of CPMIP metrics per institution and experiment, including a detailed analysis and discussion of each of the measurements. During the analysis, we found a positive correlation between the core hours needed, the complexity of the models, and the resolution used. Likewise, we show that between 5 %–15 % of the execution cost is spent in the coupling between independent components, and it only gets worse by increasing the number of resources. From the data, it is clear that queue times have a great impact on the actual speed achieved and have a huge variability across different institutions, ranging from none to up to 78 % execution overhead. Furthermore, our evaluation shows that the estimated carbon footprint of running such big simulations within the IS-ENES3 consortium is 1692 t of CO 2 equivalent. As a result of the collection, we contribute to the creation of a comprehensive database for future community reference, establishing a benchmark for evaluation and facilitating the multi-model, multi-platform comparisons crucial for understanding climate modelling performance. Given the diverse range of applications, configurations, and hardware utilised, further work is required for the standardisation and formulation of general rules. The paper concludes with recommendations for future exercises aimed at addressing the encountered challenges which will facilitate more collections of a similar nature.
- 19 April 2024
- Published by Copernicus GmbH in Geoscientific Model Development
- Vol. 17 (8) , 3081-3098
Open AccessThe computational and energy cost of simulation and storage for climate science: lessons from CMIP6
The Coupled Model Intercomparison Project (CMIP) is one of the biggest international efforts aimed at better understanding the past, present, and future of climate changes in a multi-model context. A total of 21 model intercomparison projects (MIPs) were endorsed in its sixth phase (CMIP6), which included 190 different experiments that were used to simulate 40 000 years and produced around 40 PB of data in total. This paper presents the main findings obtained from the CPMIP (the Computational Performance Model Intercomparison Project), a collection of a common set of metrics, specifically designed for assessing climate model performance. These metrics were exclusively collected from the production runs of experiments used in CMIP6 and primarily from institutions within the IS-ENES3 consortium. The document presents the full set of CPMIP metrics per institution and experiment, including a detailed analysis and discussion of each of the measurements. During the analysis, we found a positive correlation between the core hours needed, the complexity of the models, and the resolution used. Likewise, we show that between 5 %–15 % of the execution cost is spent in the coupling between independent components, and it only gets worse by increasing the number of resources. From the data, it is clear that queue times have a great impact on the actual speed achieved and have a huge variability across different institutions, ranging from none to up to 78 % execution overhead. Furthermore, our evaluation shows that the estimated carbon footprint of running such big simulations within the IS-ENES3 consortium is 1692 t of CO 2 equivalent. As a result of the collection, we contribute to the creation of a comprehensive database for future community reference, establishing a benchmark for evaluation and facilitating the multi-model, multi-platform comparisons crucial for understanding climate modelling performance. Given the diverse range of applications, configurations, and hardware utilised, further work is required for the standardisation and formulation of general rules. The paper concludes with recommendations for future exercises aimed at addressing the encountered challenges which will facilitate more collections of a similar nature.
- Open Access
- 23 October 2023
- Published by Copernicus GmbH in EGUsphere
- Vol. 2023, 1-21
The computational and energy cost of simulation and storage for climate science: lessons from CMIP6
The Coupled Model Intercomparison Project (CMIP) is one of the biggest international efforts to better understand past, present and future climate changes in a multi-model context. A total of 21 Model Intercomparison Projects (MIPs) were endorsed in its 6th phase (CMIP6), which included 190 different experiments that were used to simulate 40000 years and produced around 40 PB of data in total. This paper shows the main results obtained from the collection of performance metrics done for CMIP6 (CPMIP). The document provides the list of partners involved, the CPMIP metrics per institution/model, and the approach used for the collection and the coordination behind this process. Furthermore, a section has been included to analyze the results and prove the usefulness of the metrics for the community. Moreover, we describe the main difficulties faced during the collection and propose recommendations for future exercises.
- 23 October 2023
- Published by Copernicus GmbH in EGUsphere
- Vol. 2023, 1-21
Open AccessThe computational and energy cost of simulation and storage for climate science: lessons from CMIP6
The Coupled Model Intercomparison Project (CMIP) is one of the biggest international efforts to better understand past, present and future climate changes in a multi-model context. A total of 21 Model Intercomparison Projects (MIPs) were endorsed in its 6th phase (CMIP6), which included 190 different experiments that were used to simulate 40000 years and produced around 40 PB of data in total. This paper shows the main results obtained from the collection of performance metrics done for CMIP6 (CPMIP). The document provides the list of partners involved, the CPMIP metrics per institution/model, and the approach used for the collection and the coordination behind this process. Furthermore, a section has been included to analyze the results and prove the usefulness of the metrics for the community. Moreover, we describe the main difficulties faced during the collection and propose recommendations for future exercises.
- Open Access
- 4 May 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (5)
Semi‐Automatic Tuning of Coupled Climate Models With Multiple Intrinsic Timescales: Lessons Learned From the Lorenz96 Model
The objective of this study is to evaluate the potential for History Matching (HM) to tune a climate system with multi‐scale dynamics. By considering a toy climate model, namely, the two‐scale Lorenz96 model and producing experiments in perfect‐model setting, we explore in detail how several built‐in choices need to be carefully tested. We also demonstrate the importance of introducing physical expertise in the range of parameters, a priori to running HM. Finally we revisit a classical procedure in climate model tuning, that consists of tuning the slow and fast components separately. By doing so in the Lorenz96 model, we illustrate the non‐uniqueness of plausible parameters and highlight the specificity of metrics emerging from the coupling. This paper contributes also to bridging the communities of uncertainty quantification, machine learning and climate modeling, by making connections between the terms used by each community for the same concept and presenting promising collaboration avenues that would benefit climate modeling research.
- 4 May 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (5)
Open AccessSemi‐Automatic Tuning of Coupled Climate Models With Multiple Intrinsic Timescales: Lessons Learned From the Lorenz96 Model
The objective of this study is to evaluate the potential for History Matching (HM) to tune a climate system with multi‐scale dynamics. By considering a toy climate model, namely, the two‐scale Lorenz96 model and producing experiments in perfect‐model setting, we explore in detail how several built‐in choices need to be carefully tested. We also demonstrate the importance of introducing physical expertise in the range of parameters, a priori to running HM. Finally we revisit a classical procedure in climate model tuning, that consists of tuning the slow and fast components separately. By doing so in the Lorenz96 model, we illustrate the non‐uniqueness of plausible parameters and highlight the specificity of metrics emerging from the coupling. This paper contributes also to bridging the communities of uncertainty quantification, machine learning and climate modeling, by making connections between the terms used by each community for the same concept and presenting promising collaboration avenues that would benefit climate modeling research.
- Open Access
- 14 November 2022
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 119 (47)
Are general circulation models obsolete?
Traditional general circulation models, or GCMs—that is, three-dimensional dynamical models with unresolved terms represented in equations with tunable parameters—have been a mainstay of climate research for several decades, and some of the pioneering studies have recently been recognized by a Nobel prize in Physics. Yet, there is considerable debate around their continuing role in the future. Frequently mentioned as limitations of GCMs are the structural error and uncertainty across models with different representations of unresolved scales and the fact that the models are tuned to reproduce certain aspects of the observed Earth. We consider these shortcomings in the context of a future generation of models that may address these issues through substantially higher resolution and detail, or through the use of machine learning techniques to match them better to observations, theory, and process models. It is our contention that calibration, far from being a weakness of models, is an essential element in the simulation of complex systems, and contributes to our understanding of their inner workings. Models can be calibrated to reveal both fine-scale detail and the global response to external perturbations. New methods enable us to articulate and improve the connections between the different levels of abstract representation of climate processes, and our understanding resides in an entire hierarchy of models where GCMs will continue to play a central role for the foreseeable future.
- 14 November 2022
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 119 (47)
Open AccessAre general circulation models obsolete?
Traditional general circulation models, or GCMs—that is, three-dimensional dynamical models with unresolved terms represented in equations with tunable parameters—have been a mainstay of climate research for several decades, and some of the pioneering studies have recently been recognized by a Nobel prize in Physics. Yet, there is considerable debate around their continuing role in the future. Frequently mentioned as limitations of GCMs are the structural error and uncertainty across models with different representations of unresolved scales and the fact that the models are tuned to reproduce certain aspects of the observed Earth. We consider these shortcomings in the context of a future generation of models that may address these issues through substantially higher resolution and detail, or through the use of machine learning techniques to match them better to observations, theory, and process models. It is our contention that calibration, far from being a weakness of models, is an essential element in the simulation of complex systems, and contributes to our understanding of their inner workings. Models can be calibrated to reveal both fine-scale detail and the global response to external perturbations. New methods enable us to articulate and improve the connections between the different levels of abstract representation of climate processes, and our understanding resides in an entire hierarchy of models where GCMs will continue to play a central role for the foreseeable future.
- Open Access
- 27 October 2022
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 14 (11)
Explainable Artificial Intelligence for Bayesian Neural Networks: Toward Trustworthy Predictions of Ocean Dynamics
The trustworthiness of neural networks is often challenged because they lack the ability to express uncertainty and explain their skill. This can be problematic given the increasing use of neural networks in high stakes decision‐making such as in climate change applications. We address both issues by successfully implementing a Bayesian Neural Network (BNN), where parameters are distributions rather than deterministic, and applying novel implementations of explainable AI (XAI) techniques. The uncertainty analysis from the BNN provides a comprehensive overview of the prediction more suited to practitioners' needs than predictions from a classical neural network. Using a BNN means we can calculate the entropy (i.e., uncertainty) of the predictions and determine if the probability of an outcome is statistically significant. To enhance trustworthiness, we also spatially apply the two XAI techniques of Layer‐wise Relevance Propagation (LRP) and SHapley Additive exPlanation (SHAP) values. These XAI methods reveal the extent to which the BNN is suitable and/or trustworthy. Using two techniques gives a more holistic view of BNN skill and its uncertainty, as LRP considers neural network parameters, whereas SHAP considers changes to outputs. We verify these techniques using comparison with intuition from physical theory. The differences in explanation identify potential areas where new physical theory guided studies are needed.
- 27 October 2022
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 14 (11)
Open AccessExplainable Artificial Intelligence for Bayesian Neural Networks: Toward Trustworthy Predictions of Ocean Dynamics
The trustworthiness of neural networks is often challenged because they lack the ability to express uncertainty and explain their skill. This can be problematic given the increasing use of neural networks in high stakes decision‐making such as in climate change applications. We address both issues by successfully implementing a Bayesian Neural Network (BNN), where parameters are distributions rather than deterministic, and applying novel implementations of explainable AI (XAI) techniques. The uncertainty analysis from the BNN provides a comprehensive overview of the prediction more suited to practitioners' needs than predictions from a classical neural network. Using a BNN means we can calculate the entropy (i.e., uncertainty) of the predictions and determine if the probability of an outcome is statistically significant. To enhance trustworthiness, we also spatially apply the two XAI techniques of Layer‐wise Relevance Propagation (LRP) and SHapley Additive exPlanation (SHAP) values. These XAI methods reveal the extent to which the BNN is suitable and/or trustworthy. Using two techniques gives a more holistic view of BNN skill and its uncertainty, as LRP considers neural network parameters, whereas SHAP considers changes to outputs. We verify these techniques using comparison with intuition from physical theory. The differences in explanation identify potential areas where new physical theory guided studies are needed.
- Open Access
- 26 August 2022
- Published by Wiley in ESS Open Archive
Semi-automatic tuning of coupled climate models with multiple intrinsic timescales: lessons learned from the Lorenz96 model
The objective of this study is to evaluate the potential for History Matching (HM) to tune a climate system with multi-scale dynamics. By considering a toy climate model, namely, the two-scale Lorenz96 model and producing experiments in perfect-model setting, we explore in detail how several built-in choices need to be carefully tested. We also demonstrate the importance of introducing physical expertise in the range of parameters, a priori to running HM. Finally we revisit a classical procedure in climate model tuning, that consists of tuning the slow and fast components separately. By doing so in the Lorenz96 model, we illustrate the non-uniqueness of plausible parameters and highlight the specificity of metrics emerging from the coupling. This paper contributes also to bridging the communities of uncertainty quantification, machine learning and climate modeling, by making connections between the terms used by each community for the same concept and presenting promising collaboration avenues that would benefit climate modeling research.
- 26 August 2022
- Published by Wiley in ESS Open Archive
Open AccessSemi-automatic tuning of coupled climate models with multiple intrinsic timescales: lessons learned from the Lorenz96 model
The objective of this study is to evaluate the potential for History Matching (HM) to tune a climate system with multi-scale dynamics. By considering a toy climate model, namely, the two-scale Lorenz96 model and producing experiments in perfect-model setting, we explore in detail how several built-in choices need to be carefully tested. We also demonstrate the importance of introducing physical expertise in the range of parameters, a priori to running HM. Finally we revisit a classical procedure in climate model tuning, that consists of tuning the slow and fast components separately. By doing so in the Lorenz96 model, we illustrate the non-uniqueness of plausible parameters and highlight the specificity of metrics emerging from the coupling. This paper contributes also to bridging the communities of uncertainty quantification, machine learning and climate modeling, by making connections between the terms used by each community for the same concept and presenting promising collaboration avenues that would benefit climate modeling research.
- Open Access
- 30 April 2022
- Published by Wiley in ESS Open Archive
Explainable Artificial Intelligence for Bayesian Neural Networks: Towards trustworthy predictions of ocean dynamics
The trustworthiness of neural networks is often challenged because they lack the ability to express uncertainty and explain their skill. This can be problematic given the increasing use of neural networks in high stakes decision-making such as in climate change applications. We address both issues by successfully implementing a Bayesian Neural Network (BNN), where parameters are distributions rather than deterministic, and applying novel implementations of explainable AI (XAI) techniques. The uncertainty analysis from the BNN provides a comprehensive overview of the prediction more suited to practitioners’ needs than predictions from a classical neural network. Using a BNN means we can calculate the entropy (i.e. uncertainty) of the predictions and determine if the probability of an outcome is statistically significant. To enhance trustworthiness, we also spatially apply the two XAI techniques of Layer-wise Relevance Propagation (LRP) and SHapley Additive exPlanation (SHAP) values. These XAI methods reveal the extent to which the BNN is suitable and/or trustworthy. Using two techniques gives a more holistic view of BNN skill and its uncertainty, as LRP considers neural network parameters, whereas SHAP considers changes to outputs. We verify these techniques using comparison with intuition from physical theory. The differences in explanation identify potential areas where new physical theory guided studies are needed.
- 30 April 2022
- Published by Wiley in ESS Open Archive
Open AccessExplainable Artificial Intelligence for Bayesian Neural Networks: Towards trustworthy predictions of ocean dynamics
The trustworthiness of neural networks is often challenged because they lack the ability to express uncertainty and explain their skill. This can be problematic given the increasing use of neural networks in high stakes decision-making such as in climate change applications. We address both issues by successfully implementing a Bayesian Neural Network (BNN), where parameters are distributions rather than deterministic, and applying novel implementations of explainable AI (XAI) techniques. The uncertainty analysis from the BNN provides a comprehensive overview of the prediction more suited to practitioners’ needs than predictions from a classical neural network. Using a BNN means we can calculate the entropy (i.e. uncertainty) of the predictions and determine if the probability of an outcome is statistically significant. To enhance trustworthiness, we also spatially apply the two XAI techniques of Layer-wise Relevance Propagation (LRP) and SHapley Additive exPlanation (SHAP) values. These XAI methods reveal the extent to which the BNN is suitable and/or trustworthy. Using two techniques gives a more holistic view of BNN skill and its uncertainty, as LRP considers neural network parameters, whereas SHAP considers changes to outputs. We verify these techniques using comparison with intuition from physical theory. The differences in explanation identify potential areas where new physical theory guided studies are needed.
- Open Access
- 28 March 2022
- Published by Copernicus GmbH
CPMIP: Computational evaluation of the new era of complex Earth System Models. Multi-model results from CMIP6 and challenges for the exascale computing.
The increase in Earth System Models (ESMs) capabilities is strongly linked to the amount of computing power and data storage capacity available. The scientific community requires increased model resolution, large numbers of experiments and ensembles to quantify uncertainty, increased complexity of ESMs (including additional components), and longer simulation periods compared to the current state of climate models. HPC is currently undergoing a major change as is the next generation of computing systems (‘exascale systems’). These challenges cannot be met by mere extrapolation but require radical innovation in several computing technologies and numerical algorithms. Most applications targeting exascale machines require some degree of rewriting to expose more parallelism, and many face severe strong-scaling challenges if they are to effectively progress to exascale, as is demanded by their science goals. However, the performance evaluation of the new models through the exascale path will also increase its complexity. We do need new approaches to ensure that the computational evaluation of this new generation of models is done correctly. Moreover, this evaluation will help in the computational analysis during the model’s development and ensure the maximum throughput possible in the moment that operational configurations such as CMIP are run. CPMIP metrics are a universal set of metrics easy to collect, which provide a new way to study ESMs from a computational point of view. Thanks to the H2020 project IS-ENES3, we had a unique opportunity to exploit this new set of metrics to create a novel database based on CMIP6 experiments, using the different models and platforms available all across Europe. The results and analysis are presented here, where both differences and similarities among the models can be observed on a variety of different hardware. Moreover, the current database is presented for different studies, such as the comparison of different models running similar configurations or the same model and configuration but executed on different platforms. All these possibilities create a unique context that has to be exploited by the community to improve the evaluation of the computational performance of the ESMs, using this information for future optimizations and preparing our models for the new exascale platforms. Eventually, general prescriptions on how to disseminate the work done are given, and the need for the community to undertake the use of CPMIP metrics both on actual and new generation's platform is presented.
- 28 March 2022
- Published by Copernicus GmbH
Open AccessCPMIP: Computational evaluation of the new era of complex Earth System Models. Multi-model results from CMIP6 and challenges for the exascale computing.
The increase in Earth System Models (ESMs) capabilities is strongly linked to the amount of computing power and data storage capacity available. The scientific community requires increased model resolution, large numbers of experiments and ensembles to quantify uncertainty, increased complexity of ESMs (including additional components), and longer simulation periods compared to the current state of climate models. HPC is currently undergoing a major change as is the next generation of computing systems (‘exascale systems’). These challenges cannot be met by mere extrapolation but require radical innovation in several computing technologies and numerical algorithms. Most applications targeting exascale machines require some degree of rewriting to expose more parallelism, and many face severe strong-scaling challenges if they are to effectively progress to exascale, as is demanded by their science goals. However, the performance evaluation of the new models through the exascale path will also increase its complexity. We do need new approaches to ensure that the computational evaluation of this new generation of models is done correctly. Moreover, this evaluation will help in the computational analysis during the model’s development and ensure the maximum throughput possible in the moment that operational configurations such as CMIP are run. CPMIP metrics are a universal set of metrics easy to collect, which provide a new way to study ESMs from a computational point of view. Thanks to the H2020 project IS-ENES3, we had a unique opportunity to exploit this new set of metrics to create a novel database based on CMIP6 experiments, using the different models and platforms available all across Europe. The results and analysis are presented here, where both differences and similarities among the models can be observed on a variety of different hardware. Moreover, the current database is presented for different studies, such as the comparison of different models running similar configurations or the same model and configuration but executed on different platforms. All these possibilities create a unique context that has to be exploited by the community to improve the evaluation of the computational performance of the ESMs, using this information for future optimizations and preparing our models for the new exascale platforms. Eventually, general prescriptions on how to disseminate the work done are given, and the need for the community to undertake the use of CPMIP metrics both on actual and new generation's platform is presented.
- Open Access
- 28 March 2022
- Published by Copernicus GmbH
Histroy Matching for the tuning of coupled models: experiments on the Lorenz 96 model
The process of relying on experience and intuition to find good sets of parameters, commonly referred to as "parameter tuning" keeps having a central role in the roadmaps followed by dozens of modeling groups involved in community efforts such as the Coupled Model Intercomparison Project (CMIP). In this work, we study a tool from the Uncertainty Quantification community that started recently to draw attention in climate modeling: History Matching also referred to as "Iterative Refocussing". The core idea of History Matching is to run several simulations with different set of parameters and then use observed data to rule-out any parameter settings which are "implausible". Since climate simulation models are computationally heavy and do not allow testing every possible parameter setting, we employ an emulator that can be a cheap and accurate replacement. Here a machine learning algorithm, namely, Gaussian Process Regression is used for the emulating step. History Matching is then a good example where the recent advances in machine learning can be of high interest to climate modeling.One objective of this study is to evaluate the potential for history matching to tune a climate system with multi-scale dynamics. By using a toy climate model, namely, the Lorenz 96 model, and producing experiments in perfect-model setting, we explore different types of applications of HM and highlight the strenghts and challenges of using such a technique.
- 28 March 2022
- Published by Copernicus GmbH
Open AccessHistroy Matching for the tuning of coupled models: experiments on the Lorenz 96 model
The process of relying on experience and intuition to find good sets of parameters, commonly referred to as "parameter tuning" keeps having a central role in the roadmaps followed by dozens of modeling groups involved in community efforts such as the Coupled Model Intercomparison Project (CMIP). In this work, we study a tool from the Uncertainty Quantification community that started recently to draw attention in climate modeling: History Matching also referred to as "Iterative Refocussing". The core idea of History Matching is to run several simulations with different set of parameters and then use observed data to rule-out any parameter settings which are "implausible". Since climate simulation models are computationally heavy and do not allow testing every possible parameter setting, we employ an emulator that can be a cheap and accurate replacement. Here a machine learning algorithm, namely, Gaussian Process Regression is used for the emulating step. History Matching is then a good example where the recent advances in machine learning can be of high interest to climate modeling.One objective of this study is to evaluate the potential for history matching to tune a climate system with multi-scale dynamics. By using a toy climate model, namely, the Lorenz 96 model, and producing experiments in perfect-model setting, we explore different types of applications of HM and highlight the strenghts and challenges of using such a technique.