Phillip Wulf’s Gravity Journal – Day 2

We had dinner with Arni and Freyja last evening. Our wives had decided to spend the coming day together. The word “spa” had been mentioned several times. Arni and I were scheduled for separate private tours and interviews. The girls ditched us after dinner giving Arni and I a chance to talk shop and share what we knew about LIGO.

LIGO is a telescope for looking at gravitational waves — ripples in spacetime caused by massive objects changing speed or direction.  I prefer a less formal definition given by the granddaughter of a friend of mine. “Papa, gravitational waves are like waves made by a boat far away at sea. By the time the waves reach the shore you can’t see them, but they’re still there.” Sums it up I think.

I encourage the reader to watch these interesting and helpful gravitational wave YouTube videos, Catching Gravitational Waves — Sheila Rowan and Q & A – Catching Gravitational Waves — Sheila Rowan.

Arni and I pieced together the following LIGO development history.

LIGO 1.0 — The first went online in 2002. Unfortunately, it was not sensitive enough, and operated for nine years without detecting a single gravitational wave.

LIGO 2.0 — Failure that LIGO 1.0 was, advances in laser and vacuum technology encouraged investors to pay for a massive upgrade which went online in 2016. Approximately two weeks after going online, on 11 February 2016, LIGO 2.0 captured its first gravitational wave (GW). The GW was caused by two spiraling black holes colliding 1.3 billion light years away. Ironically, the measurable GW lasted only a few seconds, so if LIGO 2.0 had gone back online two weeks and bit later, they would have missed the serendipitous event altogether. Since then LIGO 2.0 has been a tremendous success, having captured over a hundred verified GW events.

LIGO 3.0 — Several years ago a team of system architects and scientists began designing LIGO 3.0. Arni was on the “3-Team” as it had been unofficially called. Arni had been developing ultrahigh vacuum systems for a research lab-contractor, when Professor Rowan called him at his home in Sweden several weeks ago. Arni said the LIGO 3.0 upgrade will be a radical improvement.

The previous upgrade required LIGO to shutdown for years. In contrast, most of the LIGO 3.0 upgrades took place in parallel while current LIGO 2.0 stayed online. Arnie said they will shut down LIGO 2.0 in three weeks, swap in the final LIGO 3.0 upgrades, and go back online in forty-five days — mostly due to the time required for vacuum pump-down and cryogenics to complete.

Arni helped me make sense of a few research papers I had been struggling with so that I would be better prepared for my tour this morning. Both Arni and I agreed that neither of us would have had a chance at such an opportunity like this if the LIGO science and academic communities were not so open with research and data. Much of the research had been published as ePapers, using a fantastic enabling technology named Jupyter — a kind of living notebook that readers can manipulate and remix. The LIGO Open Science Center regularly publishes GW Jupyter notebooks.

My tour this morning started with the laser. The laser is the superstar of the LIGO network. The laser actually comprises a complex optical circuit network of many lasers. It seemed to me that for the rest of the day, pretty much everyone I encountered paused to reflect when anyone said or heard the word laser.

Many scientists believe the successful capture of a 1.3 billion-year-old of gravity wave in September 2015 was the discovery of the century that will dramatically affect everyone on the planet. To be honest, I don’t have the education or experience to connect those dots, but something in my gut causes me to agree. I can however connect a few seemingly inconsequential dots hidden deep within the recesses of the LIGO basements that I believe will spawn a new renaissance.

Nothing in the world could adequately provide the computer processing needs of the new LIGO system. Nothing would ever be powerful or fast enough. No one could solve the problem. Fortunately a generation of underfunded radical Millennial graduate students from CalTech and MIT didn’t know they couldn’t do it.

First they remixed commercially available 3d and inkjet printers to print page scale stackable electronic integrated circuits on thin flexible insulated film called cookie sheets.  Cookie sheet electronic components were not as fast or compact as their commercial counterparts, but they were quick and easy to make using inexpensive reusable materials, required little power, and ran cool. Best of all, a lot of circuitry could be printed on a sheet and the sheets could be stacked like pages in a magazine.

Second, they learned how to print FPGA circuits. FPGAs are arrays of generic components on switched grid networks over layers of memory whose contents connect the grid to perform specific circuit functions — enabling software to virtually draw a soft electronic circuit that can be changed as needed.

Third, they learned to print ARM compatible SoC computers based on the open source Amber project.

Finally, they learned to print arrays of Amber computers and FPGAs over more layers of memory on the same cookie sheet, and someone suffering from sleep deprivation called it a supercookie. When they connected a hundred supercookies via fiber optics and ethernet, and programmed them simply and cleverly, they saw aggregate compute power greater than the sum of the parts.

The millennials had neither the time nor genetic disposition to to give their creation a fashionable brand name. They simply called it the Grid, and published it open source on GitHub. The LIGO system Grid has tens of thousands of cookie sheets online in basement server racks around the world, and developers print new and improved cookie sheets everyday. Hardware is no longer hard.

Most of my tour-interview centered on the gravitational wave sensor arrays, control systems, and signal processing, which given that the LIGO facility is itself a large sensor, covered pretty much everything. And everything ran on the Grid. I was overwhelmed. It was not until near the end of my day-long tour that I began to understand why Professor Rowan called me.

LIGO today has greater sensitivity, wider bandwidth, and more viewing angles so that more GW events can be captured from more of our universe. LIGO Hanford and Livingston have been integrated with more observatories — not just gravitational wave, but optical and radio, both terrestrial and orbital, all to provide an unprecedented view into space. Improved sensor technology has been integrated with geological centers, and a plethora of IoT sensor fleets to provide better ways to compensate for noise and interference. Integration with data centers around the globe has given LIGO massive deep data capability. This renaissance of new ideas and technology has essentially connected our Internet to the universe, reducing my problem to one thing — mining.

My first graduate degree was in geophysics. For my first job, my colleagues and I would ping the ground with precisely located explosives and measure the return signals with arrays of sensors. The ping lasted a few milliseconds. For the months that followed we wrote and ran programs on Cray supercomputers to locate what our employers needed buried deep within the earth.

Hewlett Packard hired me years later to help build color printers. As the result of a fluke of nature, creating color maps for printers, and digital subterranean mining are very similar. From my point of view, they are the same problem. So is looking for gravitational waves in space. I think Professor Rowan figured this out.

Professor Rowan called me this evening asking if it was okay if both Jianmin and I joined a conference call with her and a few colleagues. Jianmin is not shy. She pummeled the callers with questions and they loved her for it. I think they had already had enough of me. The colleagues thanked us and said goodbye leaving just Me, Jianmin, and Professor Rowan.

“Call me Sheila,” Rowan said. “Everyone likes you Phil. So I am going to make you an offer. Jianmin, I see you built your last two homes. Do you also keep financial records and pay the bills?”

“Yes. Phillip is not so inclined. But he can do a good job if he needs to. He runs the household when I cannot.”

“Phil has a reputation for ignoring paperwork.” Sheila said. “Not unusual for people in Phil’s line of work. We are prepared to hire someone to help him, but I want to ask you first. Would you be interested in helping?”

“I think so.”, said Jianmin.

“I’m thinking part-time.” Sheila said. “Five-to-fifteen hours per week for expense reports, timesheets, project accounting, the odd thing or two. We can adequately compensate you and still save IGR money. Phil’s offer is in no way contingent on your answer.”

Sheila summarized the offers. The terms were more than reasonable. She said I could work from any location I wanted, including my home. She said that if I agreed to occasional travel, that she would do her best to make it pleasant for both Jianmin and me. Jianmin and I both agreed to the terms.

After our conference call ended, Jianmin said Freyja messaged her earlier to say Arni also accepted an offer.

An hour later, Steve our “MI5” escort as Jianmin called him, dropped by our suite. He carried a slim briefcase and towed two packages with a folding hand truck. Steve summarized a set of documents which we read and signed. He handed us each a personalized packet.

“Those are IGR badges, credit cards, and network credentials.” Steve said, looking at his watch. “They will be activated at 3:15 AM this morning. You can charge the electronics if you like.” He said, motioning toward the boxes on the hand truck.

“What’s in the boxes?” Jianmin asked.

“Standard communications kit.” said Steve. “iPhone, iPad, MacBook Pro, wireless folding keyboard, mouse, cables, spare batteries. Oh, and Fitbits 1 too. You can earn healthcare credits.”

Steve set the boxes on a table and folded the hand truck. “You are both probably tired. We can arrange for a late flight home tomorrow if you like. Arni and Freya thought 2:00 PM tomorrow would work. You?”

Jianmin and I agreed. Steve said he would confirm the pickup time with Arni and Freya, then said goodnight.

We watched The Office before turning in.


  1. Fitbit

Leave a Reply