Meet Karthik Mukundakrishnan, SRT's Director of Research and Development.
With more than 15 years of experience leading development teams working on HPC codes for simulation software with 9 of those at Stone Ridge Technology (SRT), Dr. Karthik Mukundakrishnan is our next candidate for the SRT spotlight. Karthik currently serves as SRT’s Director of Research and Development and joined SRT in May 2013 from Dassault Systemes Simulia Inc (formerly, ABAQUS). He holds a bachelor's degree in Mechanical Engineering from the College of Engineering, Guindy (India) and obtained his M.S.E and Ph.D. in Mechanical Engineering and Applied Mechanics from the University of Pennsylvania, Philadelphia.
Karthik, you have worked on the development of ECHELON’s GPU reservoir simulation code for close to 9 years now, tell me about the early part of the journey developing ECHELON.
When I started working for SRT, not many engineering simulation companies were looking into the potential of GPUs, with the exception of a few seismic image processing teams in the oil and gas industry. One of the reasons for many companies to shy away from using GPUs was a prevailing myth that GPUs were primarily suited for performing embarrassingly parallel operations (such as vector addition) or regular stencil-based computations on highly structured grids/meshes.
So, I was happy to be joining an elite team of computational scientists attempting to simulate a complex physical process, namely, subsurface multiphase flow simulation or reservoir simulation, entirely on the GPUs. It was a challenge I wholeheartedly embraced.
How have things changed as the team grew?
The team (Eni and SRT) designing and developing ECHELON today are truly multi-disciplinary and I have learnt through experience that this is the right way forward to do modern engineering simulation software development.
Our development team now spans multiple continents, and so we face challenges in terms of training new developers and coordinating the development process to deliver a superior product. Careful project and personnel management play a critical role in addressing these issues. My role has expanded from being a developer of simulation code to managing the overall process, coordinating project deliveries with a large talented team, and addressing customer needs on time.
Tell me about the Consortium idea, announced at the end of last year
The main premise of the Consortium is to address the computational needs and challenges of both traditional hydrocarbon recovery and next generation energy transition requirements. Notable challenges are the computation of complex transport processes in the presence of a large uncertainty parameter space, and tackling higher fidelity real-world models using minimal hardware resources while achieving faster turnaround times. ECHELON is the vanguard of codes designed from the ground up specifically for massive parallelism. It is years ahead of the competition in terms of performance, model size capability, and scalability considering industry-standard formulations and solution convergence criteria. This is also taking into account all the many complex features and logic needed for simulating real-field reservoirs. Our efficient design and careful memory usage allow us to do far more with less hardware. The success of our four-year collaboration with Eni and the subsequent formation of the ECHELON Consortium have proven the capabilities of ECHELON to efficiently handle even the most challenging real-field models in the presence of complex field management strategies. In short, ECHELON is uniquely positioned to address the next generation challenges faced by the industry. The Consortium offers a unique multi-disciplinary cooperative opportunity for energy companies to participate in the development of new and bespoke features to address these challenges and benefit the industry at large.
In short, ECHELON is uniquely positioned to address the next generation challenges faced by the industry.
In addition to supermajors like Eni, SRT has smaller energy company clients too, why do you think they are keen to work with us?
The HPC landscape has been hugely democratized by the emergence of GPUs for general purpose computing. We want the entire industry and not just the supermajors to benefit from what we have to offer for reservoir simulation using GPUs and cloud technologies. ECHELON can be run efficiently on platforms ranging from a simple laptop equipped with an inexpensive gaming NVIDIA GPU card to a massive supercomputer or major cloud service providers such as AWS, MS Azure that have thousands of compute nodes with multiple GPUs per node. Beginning work with us is as simple as requesting a license and working with an SRT engineer. We, as a team, also take immense pride in providing diligent and responsive service to clients for them to experience the full potential of what ECHELON has to offer!
Enhanced with the help of our future Consortium members, SRT and Eni will continue to build the best reservoir simulation software in the world for any energy company to utilize.
Looking ahead, how do you see ECHELON and subsurface flow simulation software in general supporting the energy transition?
As with any transition, energy companies and society as a whole are faced with a dual challenge, namely, (i) addressing the growing energy demands that can be predominantly met only by oil and gas for the near term and, (ii) exploring ways to reduce carbon emissions from fossil fuels, and gradually increase the contribution of renewable and alternative sources of energy such as solar, wind, hydrogen, and geothermal.
Simulation tools should also be capable of addressing the transition from traditional hydrocarbon reservoir simulation to the broader modeling of subsurface transport processes relevant to emerging processes including CCUs, seasonal hydrogen storage and geothermal energy recovery.
I strongly believe that ECHELON is the only simulator that has the capability to adapt not only to the challenges posed by new workflows, but also to the rapidly evolving computational landscape. The Consortium serves as a great platform towards bringing the best energy engineers and computational scientists under one roof to set a new path or trend in meeting future challenges.
What items do you have on your desk?
Yes - the all too familiar blend of work and family life in the home office! How have you found working from home these past couple of years with a young child?
My wife always worked from home, so first of all I have to mention how little I appreciated the challenges she faced in that setup! Simply put, I have learned to work efficiently. I now juggle time with my son and blocks of time to focus on my work in the same way my wife always has.
Emily Fox is Stone Ridge Technology's Director of Communication.
What we are doing to help improve the reservoir simulation industry.
Leveraging modern GPUs and neural computing libraries, scientists are using deep learning technology to solve differential equations that emerge from a diverse set of physics-based problems. This blog introduces the topic and explores potential applications and limitations of the technology.Read article →