Mars as a Service: Cloud Computing for the Red Planet Exploration Era
What do clouds and a distant red planet with a thin atmosphere have in common? Dr. Jose Luis Vazquez-Poletti from Universidad Complutense de Madrid explains how cloud computing is being deployed in innovative space missions that take aim at Mars. He reports on the outcome of a meeting of the Mars MetNet Mission, which was held at the Finnish Meteorological Institute headquarters in Helsinki and describes in detail some of the cutting-edge research that is making use of cloud-based resources to handle the massive data expected.
The MetNet project aims to go where no other Mars missions have gone before, at least in terms of the way it will gather and then process data. This mission to Mars will be based on the power of a new type of dandelion seed-shaped, semi-hard landing vehicle called the MetNet Lander.
The leaders of the mission hope to deploy several of these oddly-shaped landers (as shown to the left) on Martian soil. While these lofty goals will take shape over a number of years, the first step in the mission to launch a MetNet Mars precursor mission with the first few landers being deployed in the coming year.
The main idea behind these vehicles is that by using a state-of-the-art inflatable entry and descent systems (instead of rigid heat shields and parachutes like those from the earlier semi-hard landing devices) the ratio of the payload mass to the overall mass is optimized. This means that more mass and volume resources are spared for the science payload.
The scientific payload of the Mars MetNet Mission encompasses separate instrument packages for the Martian surface operation phase. At the Martian surface, the lander will take panoramic pictures and will also perform observations of pressure, temperature, humidity, magnetism, as well as atmospheric optical depth.
The network of MetNet landers will provide valuable scientific data, decisive for studying the Martian atmosphere and its phenomena. Countries involved are Finland (Finnish Meteorological Institute), Russia (Lavochkin Space Association and Russian Space Institute) and Spain (Instituto Nacional de Técnica Aerospacial).
The collaboration developed in Mars MetNet by our group, the Distributed Systems Architecture Research Group led by Prof. Ignacio M. Llorente from the Universidad Complutense de Madrid has much to do with cloud computing… in fact, the collaborative effort is dedicated to using cloud computing for boosting all possible applications pertaining to the Mars mission, as will be explained in greater detail in a moment.
We began this collaboration with the Mars MetNet Mission more than a year ago, when they were dealing with the tracing of Phobos, the biggest Martian moon which orbits at about 9,400 Km (5,800 miles) distance from the planet’s center, completing its cycle nearly 3 times a day (a Martian day lasts 24:39 hours).
The prediction of each Phobos’ eclipse is important for the onboard instruments, which obviously depend on the landing coordinates. The challenge arises when the approximated landing area is not known until two hours before the touchdown. For this reason, an application for tracing Phobos was developed by the Meiga-MetNet Team, in order to provide a Phobos cyclogram, which is the trajectory of the Martian moon in Astronomy terms, using coordinates, dates and time intervals as an input. This way, the MetNet lander would achieve its exact location on the Martian surface by comparing the position of Phobos and the cyclogram, that is to be sent to the probe before the landing procedure.
We performed an initial parallelization of the application so that the complete set of coordinates pertaining to the approximated landing area can be processed with a desired grain. This process of profiling brought us to the conclusion that the needed hardware could be too expensive for executing this HPC application only twice a year. We had no way of even knowing if there would be other uses for this costly hardware either.
For this reason we turned to Amazon EC2, the de facto standard public cloud, attracted by its high speed deployment and its “pay-as-you-go” basis. Because all the possible setups that Amazon EC2 was offering by means of instance types and number, we crafted and validated an execution model for the application considering time, cost and a metric involving both . This way, the optimal infrastructure could be obtained given a problem size.
Considering one of the possible setups, its baremetal equivalent could be a cluster consisting in 37 nodes of the latest HP Proliant DL170 G6 Server (for example). Taking its web price of $4,909 per node, we would get our machines for $181,633 without considering any other expenses like shipping or insurances. Great, but… what about electricity? Administrator’s salary? Startup time? Even more, are we going to use this infrastructure at full power in a 24×7 fashion? Probably not.
On the other hand and according to our model, Amazon EC2 provides the needed infrastructure for $7.50.
During the meeting, I performed a comprehensive presentation explaining what Cloud Computing is and its elements to the rest of the Mars MetNet Scientific Team. The best way to make a base scientist understand Cloud Computing is to provide a good assortment of working examples and success stories. Of course, I recommended “HPC in the Cloud” as one of the main sources of news about our favorite technology.
Among these examples was the NASA case. They begun with the Nebula initiative in 2009, providing an alternative to the costly construction of additional data centers whenever NASA scientist or engineers require additional processing. This is accomplished in a fancy way and, in my point of view, following a real life “on demand” definition, as truck containers are delivered to the demanding research centers. These shipping containers can hold up to 15,000 CPU cores or 15 petabytes of storage while proving 50% more energy efficient than traditional data centers.
However, NASA decided last December 2010 to make another step on its Cloud path: they started to use the Amazon public cloud for its ATHLETE (All-Terrain Hex-Limbed Extra-Terrestrial Explorer), to be commissioned to future Mars exploring Missions. Machine instances from Amazon EC2 are used for processing satellite high-definition images in order to take navigation decisions.
But one year before NASA, the Mars MetNet Mission was already using Amazon EC2 as I explained at the beginning of this article. The results obtained for the locations of the different Martian probes were presented during the meeting and the detection of eclipses was confirmed by the experimental (and historic) data retrieved. This confirmed that the Phobos tracing model will help the Mars MetNet Mission and that Cloud Computing will be an indispensable tool, due to the huge amount of computational power needed in a very short term of time.
After my presentation, a new application was proposed. This time it has to do with the process of the meteorological data from landers pertaining to previous Mars Missions. This work has much to do with what could be addressed as “Archaeological Computing”, because much of the raw data is about 30 years old! Despite its age, the meteorological information obtained from the landers will be very useful for the Mars MetNet Mission.
The amount of data is huge and parallelization may solve some of the problems, considering several processes which respond to certain parameters. These parameters are provided by a Meteorological Model, developed within the Finnish Meteorological Institute. However, computing resource availability is another thing to take into account regarding the numerous application executions needed, so this is where a public cloud infrastructure helps reaching the goal.
But the advantages are more, because the final framework is intended to be used with the data obtained by the Mars MetNet probe, and it will be increased while more probes from the meteorological network become part of the Martian landscape.
To conclude, space missions are bringing many HPC challenges and adopting cloud computing is a decisive move for meeting them. Additionally, all research done on cloud computing for fulfilling the space mission’s demands will revert in other areas, as other achievements outside computing already did–like lyophilized food or Velcro straps.
If you are curious about the landing procedure of these dandelion-seed-shaped landers, I really encourage you to visit the Mars MetNet Mission website and watch the animation.
About the Author
Dr. Jose Luis Vazquez-Poletti is Assistant Professor in Computer Architecture at Universidad Complutense de Madrid (Spain), and a Cloud Computing Researcher. He is (and has been) directly involved in EU funded projects, such as EGEE (Grid Computing) and 4CaaSt (PaaS Cloud), as well as many Spanish national initiatives. His interests lie mainly in how the Cloud benefits real life applications, specially those pertaining to the High Performance Computing domain.
Dr. Vazquez-Poletti is also the author of a popular article that appeared in HPC in the Cloud describing a range of upcoming cloud computing research projects pending in Europe.
 J. L. Vázquez-Poletti, G. Barderas, I. M. Llorente and P. Romero: A Model for Efficient Onboard Actualization of an Instrumental Cyclogram for the Mars MetNet Mission on a Public Cloud Infrastructure. PARA2010: State of the Art in Scientific and Parallel Computing, Reykjavík (Iceland), June 2010. Proceedings to appear in Lecture Notes in Computer Science (LNCS).