Home

Back in 1985, computing power was abysmal. Solving even a modest 2D flow problem was hard. It wasn’t, therefore, a surprise when scientists at Bangalore’s NationaI Aeronautical Laboratory (NAL) met their new director, Professor Roddam Narasimha, to ask for a faster computer.

Narasimha, arguably India’s greatest scientist in fluid mechanics, of course appreciated their predicament. But there wasn’t much he could do. Computers were horrendously expensive, and even if money could be found to buy a Cray supercomputer, there was no way the Americans were going to sell it.

BuildAParallelComputer“We can’t buy a faster computer”, Narasimha told his colleagues, “but we could, like they recently did at a lab in Caltech, try to build a parallel computer to speed things up at an affordable cost”.

It was a mad and romantic idea, and it required someone equally mad to take it on. That mad scientist would be Dr Uday Narayan Sinha, then aged 39.

Sinha – henceforth “Sinhaji” in his narrative – fitted the bill. He had a shock of unkempt hair, a beard that needed urgent attention, a jhola sagging under the weight of books and journals, worn out Hawaii chappals, and a Ph.D. from IIT Kanpur for solving some absurdly difficult mathematical problem.

The fairy tale begins at this point. By the end of 1986, Sinhaji and his team actually built that parallel computer – which was named ‘Flosolver’.

PresentationToPMIt would be India’s first parallel computer even though many of us today think that C-DAC got there first. In fact, now confident after the Flosolver success, Narasimha made a presentation to Prime Minister Rajiv Gandhi explaining why we need parallel computing. I have no doubt that this heralded the birth of C-DAC.

But we mustn’t digress! This is Sinhaji story. Starting with the Flosolver success, Sinhaji blazed a trail of incredible achievement over his next two decades at NAL. He worked with tireless energy and passion: he was writing thousands of lines of code, he was building innovative hardware devices, he was parallelizing very difficult sequential code to enhance performance (there is no magical black box that accepts sequential code and spits out its optimized parallel counterpart), and he was mentoring hundreds of engineering students! It would be impossible to count how many lives he touched, and how much knowledge he imparted … but Sinhaji influenced a whole generation of computer science students now spread all over the world.

Flosolver itself became bigger, stronger and more versatile. Flow problems that appeared to be beyond the realm of the possible were easily and accurately solved.

It was my privilege to watch this remarkable karmayogi in action till 2006 (when I left NAL). He worked with the ferocious passion of a possessed spirit. He wanted absolutely to make the difference; to transcend from just clever computer science and fluid mechanics to something that could change the course of his country and the life of his countrymen.

What could be this game-changing endeavor? A horrific cyclone in Odisha during the late 1990s provided him the answer. His life’s goal henceforth would be to improve India’s capability in numerical weather prediction. In his idyllic world, no cyclone, avalanche or cloud-burst would be allowed to kill innocent Indians!

The funny thing is that the dynamical equations that help us predict weather are well-known … have been well-known for a long, long time. But they require immense computing power (hence parallel processing) and exceptional mathematical and modeling acumen. All this is very, very hard.

VarshaForecastThis would be Sinhaji’s challenge. Starting with the GCM T-80 code from NCEP, Sinhaji re-engineered, re-worked, parallelized, added more variables, understood the internal computing and modeling mechanisms, and eventually unveiled the Varsha software for numerical weather prediction over India.

Varsha has been around for over a decade now. For many years NAL’s Flosolver Lab published a daily India rainfall forecast during the monsoon months (it couldn’t be disseminated online those days because NAL couldn’t be India’s official forecaster). Varsha performed reasonably well; it could, for example, spot the horrendous Mumbai rainfall one day in July 2005.

But, in 2013, Varsha‘s prediction of the July rainfall was significantly off the mark. Sinhaji could’ve dismissed it as a rare blip, but he knew that something was wrong. So he agonized for many months to unravel the glitch – and, while doing so, stumbled on an idea that could eventually be of great consequence.

Today, Varsha is better than ever before, but it can obviously get even better. Improving it will continue to be Sinhaji’s life goal.

I often wondered what drove Sinhaji to tirelessly strive and stretch his arms towards that elusive perfection. One day I even asked him why he was doing all this although he was a grand retired gentleman over 65. “Oh, I’m merely doing my dharma”, he replied very matter-of-factly.

What are the tenets of Sinhaji’s dharma anyway? First, as we said before, that he should do something useful for his country. His big dream is that all his work will one day save at least one Indian life. He is shocked and enraged at the country’s inability to combat natural disasters. “When a cyclone need not kill anyone, why should hundreds die?”, he asks.

Second, Sinhaji’s dharma requires that he pursue excellence without wavering, and at every step. This pursuit took him to difficult problems in mathematics and fluid mechanics, and indeed even to the beautiful world of Indian verse and music. He still finds Indian poetry deeply thoughtful and enchanting, and his favorite pastime is still to play a mellifluous tune on his violin.

Third, Sinhaji’s dharma is to be truthful, unflinchingly and at all times. As someone who occasionally chronicled Flosolver’s achievement, I often asked him if I should report an unfavorable result, or just keep quiet about it. He always advised me to be completely truthful. In fact, his experiments with truth reveal the overwhelming influence of Mahatma Gandhi.

Finally, Sinhaji’s dharma requires him to be extremely aggressive in the pursuit of his goals. At most times he is extremely amiable, even childlike. But if you become an infringement in one of his many moral or scientific experiments, you will be badly hurt. “My hero is Parshuram. Like him, I won’t hesitate to harm anyone who strays from the righteous path”, he often tells you.

Now that I’m an independent freelancer, I’ve made a conscious attempt to reach out to Sinhaji in 2017. Even after all these years he hasn’t changed much. The flesh is a little weaker, but the spirit seems even stronger. It is still that same Sinhaji offering you a cup of lemon tea, or inviting you to join him as he potters around his beloved books. He still  lovingly browses through the “Complete Works of Euler”. If he is pressed for time, he moves away after bestowing the book with such a loving gaze that the book might well blush.

At other times, when he is seized by the mathematical bug while driving for some meeting, he will pull out his pen from his pocket, grab his car driver’s newspaper, and scribble a completely new proof to show that the square root of three is irrational. Or explain how, when everything else fails, randomness could prove to be the key. Or even ruminate on artificial intelligence and worry that India remains unprepared for the new revolution.

Sinhaji’s complete and unequivocal faith in the power of science, that allows him to boldly venture into unknown technological terrain, never wavers. “Science never lets you down, only we let science down”, he likes telling you.

CrayAwardAs an long-time admirer of Sinhaji (and there are hundreds like me!), I’m bothered that we may be letting Sinhaji down. Sinhaji hasn’t been recognized enough (although he did recently get Cray’s A P J Abdul Kalam Award) for his phenomenal achievement and body of work. But you know how it tends to be: awards only come to those who want them, or ask for him, or make sure they ‘belong’ to those enabling cliques. Instead, Sinhaji would be happier if he got more opportunities to serve his people – it was the same admirable resolve, half a century ago, that saw a young Sinhaji travel from village to village in Bihar during a virulent outbreak of cholera  hydrating young children and possibly saving their lives.

Today Sinhaji turned 70, but he soldiers on. The tapasya continues. And he asks for nothing in return. After all, he’s only doing his dharma.

 

5 thoughts on “Why don’t more Indians know U N Sinha?

  1. Dear Srinivas, Thanks for the post. Please do convey my regards to him even though Dr. Sinha may not remember me though I vividly remember when way back he proudly showed his Flosolver to us visiting science journalists NAL and we wrote about it.
    Bajji.

  2. Great narrative BM! One does hope that this karmayogi gets his due recognition. Maybe those who were touched by him can take it as their dharma to ensure that he does.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s