Sport Optimised / Ars Technica Feature
F1 drivers practice similar g-force to Apollo astronauts during Earth re-entry. Here’s how they design and make the cars.
For over sixty years, Formula one teams have developed, tested, and built the fastest and most technologically epic cars the world has ever seen. An almost unending list of superlatives can be ladled onto F1 cars: they can accelerate from zero to 190mph in about ten seconds, fling around a corner at such speeds that the driver practices g-force close to that of an Apollo astronaut during Earth re-entry, and then decelerate by 60mph in just 0.7 seconds thanks to strong brakes and massive downforce—the same downforce that stopped the car from spinning out around that corner.
But the bit that’s indeed amazing is that these machines are designed and built from scrape every year. That’s what makes F1 so competitive and why the rate of improvement is so rapid. These teams—there are only about ten of them, and most are based in England—have been challenging each other to make a fresh best-car-in-the-world every year for sixty years. The only way to pole position is to attempt to find an edge that no one else has thought of yet and then to keep finding fresh edges when everyone inevitably catches up.
As you’ve very likely guessed, materials science, engineering, bleeding-edge software, and recently the cloud are a major part of F1 innovation—and indeed, those fair topics are where we lay our scene.
For this story I embedded with Renault Sport Formula One Team as they made their final preparations for the two thousand seventeen season. As I write this, I can hear this year’s cars being tested around Circuit de Barcelona-Catalunya; a Mercedes car has just set the fastest lap time, and we’re all silently wondering if they will predominate once again.
After a difficult 2016, things are looking up for Renault Sport Formula One Team in 2017. They’re back with a fresh chassis and a fresh, fully integrated Renault power unit. The engineering teams have been reinforced with fresh recruits and the acquisition of state-of-the-art tooling and machines. Planning, design, and international collaboration and communications have been bolstered with a renewed partnership with Microsoft Cloud. And F1 legend Alain Prost is on board to advise drivers Nico Hülkenberg and Jolyon Palmer.
How will they fare? I don’t know; I’m a tech journalist, not a motorsports correspondent. But I can tell you how they built that car—or more accurately, how they built and scrapped thousands of possible, prototype cars in their search for one championship-winning design.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
The Brabham BT46 “fan car.”
A lot of switches were made to F1 cars following Senna’s death, including the mandatory wooden plank (skid block) which shows if a car is running too low to the ground (and thus cracking regulations).
The discovery of downforce
For the very first thirty years of Formula 1’s history, the cars were mostly dumb mechanical animals; not much mattered beyond the driver, tyres, and power train. Then, in 1977, Team Lotus (fairly different from the latest Lotus F1 team, which then became Renault Sport Formula One Team) began paying more attention to aerodynamics—specifically the ground effect, which, in the world of motorsport, is usually known as downforce. The underside of the Lotus seventy nine F1 car was curved like an upside-down airplane wing, creating a pocket of low pressure that essentially sucked the car to the ground.
The Lotus seventy nine was massively successful, and before long—once the other teams eventually sussed out Lotus’ black magic—every Formula one car was sculpted to provide maximum downforce. One design, the Brabham BT46 (pictured above), even had a big ol’ fan that sucked air out from underneath the car.
Over the next few years Formula one got swifter and quicker, especially around corners. Eventually, following a number of accidents and the death of Gilles Villeneuve in 1982, the FIA mandated a come back to flat-bottomed cars. The aerodynamic cat couldn’t be put back in the bag, tho’.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
A 60-percent scale model of an F1 car inwards the wind tunnel.
Renault Sport Formula One Team
Formula one teams also have very detailed car simulators to augment the limited amount of live track testing.
Renault Sport Formula One Team
The car simulator control room.
An F1 car within a computational fluid dynamics (CFD) simulation.
Renault Sport Formula One Team
A CFD aerodynamics engineer.
Renault Sport Formula One Team
The supercomputer that runs the F1 team’s CFD simulations. The team has a collaboration with Boeing, thus the Boeing label.
Renault Sport Formula One Team
Another nice shot of the CFD supercomputer.
25 teraflops and not a drop more
Almost every area of technological and engineering advancement in F1 has followed a similar path to aerodynamics. A team finds an area that hasn’t yet been regulated by the FIA, or where existing regulation can be creatively interpreted; that team thrusts to within a few millimetres of the regulations, sometimes stepping slightly over the line; other teams go after suit; then the FIA revises its regulations and the cycle starts again.
As you can imagine, then, after some sixty years of attempting to outwit the feds, Formula one today is governed by a rather long list of regulations—hundreds of pages of them, in fact.
For example, each Formula one team is only permitted to use twenty five teraflops (trillions of floating point operations per 2nd) of dual precision (64-bit) computing power for simulating car aerodynamics. Twenty five teraflops isn’t a lot of processing power, in the grand scheme of supercomputers: it’s about comparable to twenty five of the original Nvidia Titan graphics cards (the fresh Pascal-based cards are no good at double-precision maths).
Oddly, the F1 regulations also stipulate that only CPUs can be used, not GPUs, and that teams must explicitly prove whether they’re using AVX instructions or not. Without AVX, the FIA rates a single Sandy Bridge or Ivy Bridge CPU core at four flops; with AVX, each core is rated at eight flops. Every team has to submit the exact specifications of their compute cluster to the FIA at the begin of the season, and then a logfile after every eight weeks of ongoing testing.
Renault Sport Formula One Team recently deployed a fresh on-premises compute cluster with Legitimate,000 cores—so, very likely about Two,000 Intel Xeon CPUs. While the total number of teraflops is rigorously limited, other aspects of the system’s architecture can be optimised. For example, the team’s cluster has very parallel storage. “Each compute knot has a dedicated connection to storage so that we don’t waste flops on reading and writing data,” says Mark Everest, one of the team’s infrastructure managers. “There was a big improvement in spectacle when we switched from our old cluster to the fresh one, without necessarily switching the software,” and with the same 25-teraflops processing cap, Everest adds.
Everest says that every team has its own on-premises hardware setup and that no one has yet moved to the cloud. There’s no technical reason why the cloud can’t be used for car aerodynamics simulations—and F1 teams are investigating such a possibility—but the aforementioned stringent CPU stipulations presently make it unlikely. The result is that most F1 teams use a somewhat hybridised setup, with a local Linux cluster outputting aerodynamics data that informs the manufacturing of physical components, the details of which are kept in the cloud.
Making Formula one more titillating
The cars this year will have broader tyres providing more grip and broader wings generating more downforce. Lap times will be diminished significantly as drivers fling themselves around corners at speeds not seen since the turn of the century. It’s hard to say whether the racing will actually be more arousing; generally, extra downforce isn’t a good thing, and the extra width of the cars might make it firmer to overtake.
Wind tunnel usage is similarly restricted: F1 teams are only permitted twenty five hours of “wind on” time per week to test fresh chassis designs. Ten years ago, in 2007, it was very different, says Everest: “There was no limitation on teraflops, no confinement on wind tunnel hours,” resumes Everest. “We had three shifts running the wind tunnel 24/7. It got to the point where a lot of teams were talking about building a 2nd wind tunnel; Williams built a 2nd tunnel.
“We determined to go down the computing route, with CFD—computational fluid dynamics—rather than build another wind tunnel. When we built our fresh compute cluster in 2007, the plan was that we’d dual our compute every year. Very quickly it was realised that the teams with meaty budgets—the manufacturer-backed teams—would get an unfair advantage over smaller teams, because they didn’t have the money to build these enormous clusters.”
Soon after, to prevent the larger F1 teams from throwing more and more money at aerodynamics, the FIA began restricting both wind tunnel usage and compute power for simulations.
Formula 1: A technical deep dive into building the world’s fastest cars, Ars Technica
Sport Optimised / Ars Technica Feature
F1 drivers practice similar g-force to Apollo astronauts during Earth re-entry. Here’s how they design and make the cars.
For over sixty years, Formula one teams have developed, tested, and built the fastest and most technologically incredible cars the world has ever seen. An almost unending list of superlatives can be ladled onto F1 cars: they can accelerate from zero to 190mph in about ten seconds, fling around a corner at such speeds that the driver practices g-force close to that of an Apollo astronaut during Earth re-entry, and then decelerate by 60mph in just 0.7 seconds thanks to strong brakes and massive downforce—the same downforce that stopped the car from spinning out around that corner.
But the bit that’s truly epic is that these machines are designed and built from scrape every year. That’s what makes F1 so competitive and why the rate of improvement is so rapid. These teams—there are only about ten of them, and most are based in England—have been challenging each other to make a fresh best-car-in-the-world every year for sixty years. The only way to pole position is to attempt to find an edge that no one else has thought of yet and then to keep finding fresh edges when everyone inevitably catches up.
As you’ve very likely guessed, materials science, engineering, bleeding-edge software, and recently the cloud are a major part of F1 innovation—and indeed, those fair topics are where we lay our scene.
For this story I embedded with Renault Sport Formula One Team as they made their final preparations for the two thousand seventeen season. As I write this, I can hear this year’s cars being tested around Circuit de Barcelona-Catalunya; a Mercedes car has just set the fastest lap time, and we’re all silently wondering if they will predominate once again.
After a difficult 2016, things are looking up for Renault Sport Formula One Team in 2017. They’re back with a fresh chassis and a fresh, fully integrated Renault power unit. The engineering teams have been reinforced with fresh recruits and the acquisition of state-of-the-art tooling and machines. Planning, design, and international collaboration and communications have been bolstered with a renewed partnership with Microsoft Cloud. And F1 legend Alain Prost is on board to advise drivers Nico Hülkenberg and Jolyon Palmer.
How will they fare? I don’t know; I’m a tech journalist, not a motorsports correspondent. But I can tell you how they built that car—or more accurately, how they built and scrapped thousands of possible, prototype cars in their search for one championship-winning design.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
The Brabham BT46 “fan car.”
A lot of switches were made to F1 cars following Senna’s death, including the mandatory wooden plank (skid block) which shows if a car is running too low to the ground (and thus violating regulations).
The discovery of downforce
For the very first thirty years of Formula 1’s history, the cars were mostly dumb mechanical brutes; not much mattered beyond the driver, tyres, and power train. Then, in 1977, Team Lotus (fairly different from the latest Lotus F1 team, which then became Renault Sport Formula One Team) began paying more attention to aerodynamics—specifically the ground effect, which, in the world of motorsport, is usually known as downforce. The underside of the Lotus seventy nine F1 car was curved like an upside-down airplane wing, creating a pocket of low pressure that essentially sucked the car to the ground.
The Lotus seventy nine was massively successful, and before long—once the other teams eventually sussed out Lotus’ black magic—every Formula one car was sculpted to provide maximum downforce. One design, the Brabham BT46 (pictured above), even had a big ol’ fan that sucked air out from underneath the car.
Over the next few years Formula one got quicker and swifter, especially around corners. Eventually, following a number of accidents and the death of Gilles Villeneuve in 1982, the FIA mandated a comeback to flat-bottomed cars. The aerodynamic cat couldn’t be put back in the bag, tho’.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
A 60-percent scale model of an F1 car inwards the wind tunnel.
Renault Sport Formula One Team
Formula one teams also have very detailed car simulators to augment the limited amount of live track testing.
Renault Sport Formula One Team
The car simulator control room.
An F1 car within a computational fluid dynamics (CFD) simulation.
Renault Sport Formula One Team
A CFD aerodynamics engineer.
Renault Sport Formula One Team
The supercomputer that runs the F1 team’s CFD simulations. The team has a collaboration with Boeing, thus the Boeing label.
Renault Sport Formula One Team
Another nice shot of the CFD supercomputer.
25 teraflops and not a drop more
Almost every area of technological and engineering advancement in F1 has followed a similar path to aerodynamics. A team finds an area that hasn’t yet been regulated by the FIA, or where existing regulation can be creatively interpreted; that team thrusts to within a few millimetres of the regulations, sometimes stepping slightly over the line; other teams go after suit; then the FIA revises its regulations and the cycle commences again.
As you can imagine, then, after some sixty years of attempting to outwit the feds, Formula one today is governed by a rather long list of regulations—hundreds of pages of them, in fact.
For example, each Formula one team is only permitted to use twenty five teraflops (trillions of floating point operations per 2nd) of dual precision (64-bit) computing power for simulating car aerodynamics. Twenty five teraflops isn’t a lot of processing power, in the grand scheme of supercomputers: it’s about comparable to twenty five of the original Nvidia Titan graphics cards (the fresh Pascal-based cards are no good at double-precision maths).
Oddly, the F1 regulations also stipulate that only CPUs can be used, not GPUs, and that teams must explicitly prove whether they’re using AVX instructions or not. Without AVX, the FIA rates a single Sandy Bridge or Ivy Bridge CPU core at four flops; with AVX, each core is rated at eight flops. Every team has to submit the exact specifications of their compute cluster to the FIA at the embark of the season, and then a logfile after every eight weeks of ongoing testing.
Renault Sport Formula One Team recently deployed a fresh on-premises compute cluster with Legal,000 cores—so, most likely about Two,000 Intel Xeon CPUs. While the total number of teraflops is rigorously limited, other aspects of the system’s architecture can be optimised. For example, the team’s cluster has very parallel storage. “Each compute knot has a dedicated connection to storage so that we don’t waste flops on reading and writing data,” says Mark Everest, one of the team’s infrastructure managers. “There was a big improvement in spectacle when we switched from our old cluster to the fresh one, without necessarily switching the software,” and with the same 25-teraflops processing cap, Everest adds.
Everest says that every team has its own on-premises hardware setup and that no one has yet moved to the cloud. There’s no technical reason why the cloud can’t be used for car aerodynamics simulations—and F1 teams are investigating such a possibility—but the aforementioned stringent CPU stipulations presently make it unlikely. The result is that most F1 teams use a somewhat hybridised setup, with a local Linux cluster outputting aerodynamics data that informs the manufacturing of physical components, the details of which are kept in the cloud.
Making Formula one more titillating
The cars this year will have broader tyres providing more grip and broader wings generating more downforce. Lap times will be diminished significantly as drivers fling themselves around corners at speeds not seen since the turn of the century. It’s hard to say whether the racing will actually be more titillating; generally, extra downforce isn’t a good thing, and the extra width of the cars might make it firmer to overtake.
Wind tunnel usage is similarly restricted: F1 teams are only permitted twenty five hours of “wind on” time per week to test fresh chassis designs. Ten years ago, in 2007, it was very different, says Everest: “There was no limitation on teraflops, no limitation on wind tunnel hours,” resumes Everest. “We had three shifts running the wind tunnel 24/7. It got to the point where a lot of teams were talking about building a 2nd wind tunnel; Williams built a 2nd tunnel.
“We determined to go down the computing route, with CFD—computational fluid dynamics—rather than build another wind tunnel. When we built our fresh compute cluster in 2007, the plan was that we’d dual our compute every year. Very quickly it was realised that the teams with meaty budgets—the manufacturer-backed teams—would get an unfair advantage over smaller teams, because they didn’t have the money to build these enormous clusters.”
Soon after, to prevent the larger F1 teams from throwing more and more money at aerodynamics, the FIA began restricting both wind tunnel usage and compute power for simulations.
Formula 1: A technical deep dive into building the world’s fastest cars, Ars Technica
Sport Optimised / Ars Technica Feature
F1 drivers practice similar g-force to Apollo astronauts during Earth re-entry. Here’s how they design and make the cars.
For over sixty years, Formula one teams have developed, tested, and built the fastest and most technologically awesome cars the world has ever seen. An almost unending list of superlatives can be ladled onto F1 cars: they can accelerate from zero to 190mph in about ten seconds, fling around a corner at such speeds that the driver practices g-force close to that of an Apollo astronaut during Earth re-entry, and then decelerate by 60mph in just 0.7 seconds thanks to strong brakes and massive downforce—the same downforce that stopped the car from spinning out around that corner.
But the bit that’s truly epic is that these machines are designed and built from scrape every year. That’s what makes F1 so competitive and why the rate of improvement is so rapid. These teams—there are only about ten of them, and most are based in England—have been challenging each other to make a fresh best-car-in-the-world every year for sixty years. The only way to pole position is to attempt to find an edge that no one else has thought of yet and then to keep finding fresh edges when everyone inevitably catches up.
As you’ve very likely guessed, materials science, engineering, bleeding-edge software, and recently the cloud are a major part of F1 innovation—and indeed, those fair topics are where we lay our scene.
For this story I embedded with Renault Sport Formula One Team as they made their final preparations for the two thousand seventeen season. As I write this, I can hear this year’s cars being tested around Circuit de Barcelona-Catalunya; a Mercedes car has just set the fastest lap time, and we’re all silently wondering if they will predominate once again.
After a difficult 2016, things are looking up for Renault Sport Formula One Team in 2017. They’re back with a fresh chassis and a fresh, fully integrated Renault power unit. The engineering teams have been reinforced with fresh recruits and the acquisition of state-of-the-art tooling and machines. Planning, design, and international collaboration and communications have been bolstered with a renewed partnership with Microsoft Cloud. And F1 legend Alain Prost is on board to advise drivers Nico Hülkenberg and Jolyon Palmer.
How will they fare? I don’t know; I’m a tech journalist, not a motorsports correspondent. But I can tell you how they built that car—or more accurately, how they built and scrapped thousands of possible, prototype cars in their search for one championship-winning design.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
The Brabham BT46 “fan car.”
A lot of switches were made to F1 cars following Senna’s death, including the mandatory wooden plank (skid block) which shows if a car is running too low to the ground (and thus cracking regulations).
The discovery of downforce
For the very first thirty years of Formula 1’s history, the cars were mostly dumb mechanical animals; not much mattered beyond the driver, tyres, and power train. Then, in 1977, Team Lotus (fairly different from the latest Lotus F1 team, which then became Renault Sport Formula One Team) began paying more attention to aerodynamics—specifically the ground effect, which, in the world of motorsport, is usually known as downforce. The underside of the Lotus seventy nine F1 car was curved like an upside-down airplane wing, creating a pocket of low pressure that essentially sucked the car to the ground.
The Lotus seventy nine was massively successful, and before long—once the other teams eventually sussed out Lotus’ black magic—every Formula one car was sculpted to provide maximum downforce. One design, the Brabham BT46 (pictured above), even had a big ol’ fan that sucked air out from underneath the car.
Over the next few years Formula one got quicker and quicker, especially around corners. Eventually, following a number of accidents and the death of Gilles Villeneuve in 1982, the FIA mandated a comeback to flat-bottomed cars. The aerodynamic cat couldn’t be put back in the bag, however.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
A 60-percent scale model of an F1 car inwards the wind tunnel.
Renault Sport Formula One Team
Formula one teams also have very detailed car simulators to augment the limited amount of live track testing.
Renault Sport Formula One Team
The car simulator control room.
An F1 car within a computational fluid dynamics (CFD) simulation.
Renault Sport Formula One Team
A CFD aerodynamics engineer.
Renault Sport Formula One Team
The supercomputer that runs the F1 team’s CFD simulations. The team has a collaboration with Boeing, thus the Boeing label.
Renault Sport Formula One Team
Another nice shot of the CFD supercomputer.
25 teraflops and not a drop more
Almost every area of technological and engineering advancement in F1 has followed a similar path to aerodynamics. A team finds an area that hasn’t yet been regulated by the FIA, or where existing regulation can be creatively interpreted; that team shoves to within a few millimetres of the regulations, sometimes stepping slightly over the line; other teams go after suit; then the FIA revises its regulations and the cycle embarks again.
As you can imagine, then, after some sixty years of attempting to outwit the feds, Formula one today is governed by a rather long list of regulations—hundreds of pages of them, in fact.
For example, each Formula one team is only permitted to use twenty five teraflops (trillions of floating point operations per 2nd) of dual precision (64-bit) computing power for simulating car aerodynamics. Twenty five teraflops isn’t a lot of processing power, in the grand scheme of supercomputers: it’s about comparable to twenty five of the original Nvidia Titan graphics cards (the fresh Pascal-based cards are no good at double-precision maths).
Oddly, the F1 regulations also stipulate that only CPUs can be used, not GPUs, and that teams must explicitly prove whether they’re using AVX instructions or not. Without AVX, the FIA rates a single Sandy Bridge or Ivy Bridge CPU core at four flops; with AVX, each core is rated at eight flops. Every team has to submit the exact specifications of their compute cluster to the FIA at the begin of the season, and then a logfile after every eight weeks of ongoing testing.
Renault Sport Formula One Team recently deployed a fresh on-premises compute cluster with Eighteen,000 cores—so, most likely about Two,000 Intel Xeon CPUs. While the total number of teraflops is stringently limited, other aspects of the system’s architecture can be optimised. For example, the team’s cluster has very parallel storage. “Each compute knot has a dedicated connection to storage so that we don’t waste flops on reading and writing data,” says Mark Everest, one of the team’s infrastructure managers. “There was a big improvement in spectacle when we switched from our old cluster to the fresh one, without necessarily switching the software,” and with the same 25-teraflops processing cap, Everest adds.
Everest says that every team has its own on-premises hardware setup and that no one has yet moved to the cloud. There’s no technical reason why the cloud can’t be used for car aerodynamics simulations—and F1 teams are investigating such a possibility—but the aforementioned stringent CPU stipulations presently make it unlikely. The result is that most F1 teams use a somewhat hybridised setup, with a local Linux cluster outputting aerodynamics data that informs the manufacturing of physical components, the details of which are kept in the cloud.
Making Formula one more titillating
The cars this year will have broader tyres providing more grip and broader wings generating more downforce. Lap times will be diminished significantly as drivers fling themselves around corners at speeds not seen since the turn of the century. It’s hard to say whether the racing will actually be more titillating; generally, extra downforce isn’t a good thing, and the extra width of the cars might make it tighter to overtake.
Wind tunnel usage is similarly restricted: F1 teams are only permitted twenty five hours of “wind on” time per week to test fresh chassis designs. Ten years ago, in 2007, it was very different, says Everest: “There was no limitation on teraflops, no confinement on wind tunnel hours,” resumes Everest. “We had three shifts running the wind tunnel 24/7. It got to the point where a lot of teams were talking about building a 2nd wind tunnel; Williams built a 2nd tunnel.
“We determined to go down the computing route, with CFD—computational fluid dynamics—rather than build another wind tunnel. When we built our fresh compute cluster in 2007, the plan was that we’d dual our compute every year. Very quickly it was realised that the teams with enormous budgets—the manufacturer-backed teams—would get an unfair advantage over smaller teams, because they didn’t have the money to build these enormous clusters.”
Soon after, to prevent the larger F1 teams from throwing more and more money at aerodynamics, the FIA began restricting both wind tunnel usage and compute power for simulations.
Formula 1: A technical deep dive into building the world’s fastest cars, Ars Technica
Sport Optimised / Ars Technica Feature
F1 drivers practice similar g-force to Apollo astronauts during Earth re-entry. Here’s how they design and make the cars.
For over sixty years, Formula one teams have developed, tested, and built the fastest and most technologically outstanding cars the world has ever seen. An almost unending list of superlatives can be ladled onto F1 cars: they can accelerate from zero to 190mph in about ten seconds, fling around a corner at such speeds that the driver practices g-force close to that of an Apollo astronaut during Earth re-entry, and then decelerate by 60mph in just 0.7 seconds thanks to strong brakes and massive downforce—the same downforce that stopped the car from spinning out around that corner.
But the bit that’s indeed awesome is that these machines are designed and built from scrape every year. That’s what makes F1 so competitive and why the rate of improvement is so rapid. These teams—there are only about ten of them, and most are based in England—have been challenging each other to make a fresh best-car-in-the-world every year for sixty years. The only way to pole position is to attempt to find an edge that no one else has thought of yet and then to keep finding fresh edges when everyone inevitably catches up.
As you’ve most likely guessed, materials science, engineering, bleeding-edge software, and recently the cloud are a major part of F1 innovation—and indeed, those fair topics are where we lay our scene.
For this story I embedded with Renault Sport Formula One Team as they made their final preparations for the two thousand seventeen season. As I write this, I can hear this year’s cars being tested around Circuit de Barcelona-Catalunya; a Mercedes car has just set the fastest lap time, and we’re all silently wondering if they will predominate once again.
After a difficult 2016, things are looking up for Renault Sport Formula One Team in 2017. They’re back with a fresh chassis and a fresh, fully integrated Renault power unit. The engineering teams have been reinforced with fresh recruits and the acquisition of state-of-the-art tooling and machines. Planning, design, and international collaboration and communications have been bolstered with a renewed partnership with Microsoft Cloud. And F1 legend Alain Prost is on board to advise drivers Nico Hülkenberg and Jolyon Palmer.
How will they fare? I don’t know; I’m a tech journalist, not a motorsports correspondent. But I can tell you how they built that car—or more accurately, how they built and scrapped thousands of possible, prototype cars in their search for one championship-winning design.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
Various Ferrari Formula one cars, dating from one thousand nine hundred fifty to 2002.
The Brabham BT46 “fan car.”
A lot of switches were made to F1 cars following Senna’s death, including the mandatory wooden plank (skid block) which shows if a car is running too low to the ground (and thus cracking regulations).
The discovery of downforce
For the very first thirty years of Formula 1’s history, the cars were mostly dumb mechanical animals; not much mattered beyond the driver, tyres, and power train. Then, in 1977, Team Lotus (fairly different from the latest Lotus F1 team, which then became Renault Sport Formula One Team) embarked paying more attention to aerodynamics—specifically the ground effect, which, in the world of motorsport, is usually known as downforce. The underside of the Lotus seventy nine F1 car was curved like an upside-down airplane wing, creating a pocket of low pressure that essentially sucked the car to the ground.
The Lotus seventy nine was massively successful, and before long—once the other teams eventually sussed out Lotus’ black magic—every Formula one car was sculpted to provide maximum downforce. One design, the Brabham BT46 (pictured above), even had a big ol’ fan that sucked air out from underneath the car.
Over the next few years Formula one got quicker and swifter, especially around corners. Eventually, following a number of accidents and the death of Gilles Villeneuve in 1982, the FIA mandated a come back to flat-bottomed cars. The aerodynamic cat couldn’t be put back in the bag, tho’.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
An arty shot of a Formula one wind tunnel.
Renault Sport Formula One Team
A 60-percent scale model of an F1 car inwards the wind tunnel.
Renault Sport Formula One Team
Formula one teams also have very detailed car simulators to augment the limited amount of live track testing.
Renault Sport Formula One Team
The car simulator control room.
An F1 car within a computational fluid dynamics (CFD) simulation.
Renault Sport Formula One Team
A CFD aerodynamics engineer.
Renault Sport Formula One Team
The supercomputer that runs the F1 team’s CFD simulations. The team has a collaboration with Boeing, thus the Boeing label.
Renault Sport Formula One Team
Another nice shot of the CFD supercomputer.
25 teraflops and not a drop more
Almost every area of technological and engineering advancement in F1 has followed a similar path to aerodynamics. A team finds an area that hasn’t yet been regulated by the FIA, or where existing regulation can be creatively interpreted; that team shoves to within a few millimetres of the regulations, sometimes stepping slightly over the line; other teams go after suit; then the FIA revises its regulations and the cycle commences again.
As you can imagine, then, after some sixty years of attempting to outwit the feds, Formula one today is governed by a rather long list of regulations—hundreds of pages of them, in fact.
For example, each Formula one team is only permitted to use twenty five teraflops (trillions of floating point operations per 2nd) of dual precision (64-bit) computing power for simulating car aerodynamics. Twenty five teraflops isn’t a lot of processing power, in the grand scheme of supercomputers: it’s about comparable to twenty five of the original Nvidia Titan graphics cards (the fresh Pascal-based cards are no good at double-precision maths).
Oddly, the F1 regulations also stipulate that only CPUs can be used, not GPUs, and that teams must explicitly prove whether they’re using AVX instructions or not. Without AVX, the FIA rates a single Sandy Bridge or Ivy Bridge CPU core at four flops; with AVX, each core is rated at eight flops. Every team has to submit the exact specifications of their compute cluster to the FIA at the embark of the season, and then a logfile after every eight weeks of ongoing testing.
Renault Sport Formula One Team recently deployed a fresh on-premises compute cluster with Legal,000 cores—so, most likely about Two,000 Intel Xeon CPUs. While the total number of teraflops is stringently limited, other aspects of the system’s architecture can be optimised. For example, the team’s cluster has very parallel storage. “Each compute knot has a dedicated connection to storage so that we don’t waste flops on reading and writing data,” says Mark Everest, one of the team’s infrastructure managers. “There was a big improvement in spectacle when we switched from our old cluster to the fresh one, without necessarily switching the software,” and with the same 25-teraflops processing cap, Everest adds.
Everest says that every team has its own on-premises hardware setup and that no one has yet moved to the cloud. There’s no technical reason why the cloud can’t be used for car aerodynamics simulations—and F1 teams are investigating such a possibility—but the aforementioned stringent CPU stipulations presently make it unlikely. The result is that most F1 teams use a somewhat hybridised setup, with a local Linux cluster outputting aerodynamics data that informs the manufacturing of physical components, the details of which are kept in the cloud.
Making Formula one more arousing
The cars this year will have broader tyres providing more grip and broader wings generating more downforce. Lap times will be diminished significantly as drivers fling themselves around corners at speeds not seen since the turn of the century. It’s hard to say whether the racing will actually be more titillating; generally, extra downforce isn’t a good thing, and the extra width of the cars might make it firmer to overtake.
Wind tunnel usage is similarly restricted: F1 teams are only permitted twenty five hours of “wind on” time per week to test fresh chassis designs. Ten years ago, in 2007, it was very different, says Everest: “There was no limitation on teraflops, no limitation on wind tunnel hours,” proceeds Everest. “We had three shifts running the wind tunnel 24/7. It got to the point where a lot of teams were talking about building a 2nd wind tunnel; Williams built a 2nd tunnel.
“We determined to go down the computing route, with CFD—computational fluid dynamics—rather than build another wind tunnel. When we built our fresh compute cluster in 2007, the plan was that we’d dual our compute every year. Very quickly it was realised that the teams with ample budgets—the manufacturer-backed teams—would get an unfair advantage over smaller teams, because they didn’t have the money to build these enormous clusters.”
Soon after, to prevent the larger F1 teams from throwing more and more money at aerodynamics, the FIA began restricting both wind tunnel usage and compute power for simulations.