Spain’s IFISC Tackles COVID-19, Climate Change with GIGABYTE Servers
By using GIGABYTE, Spain’s Institute for Cross-Disciplinary Physics and Complex Systems is pitting the world’s foremost server solutions against some of the world’s most pressing issues, including the effects of climate change, pollution, and COVID-19. GIGABYTE servers are up to the diverse and daunting tasks, because they are designed for high performance computing, intensive numerical simulations, AI development, and big data management.
Few things in the world are certain, but here are a couple of cold, hard facts: one, computing power will continue to advance as industry leaders push the envelope of Moore’s Law; and two, the earth is getting hotter, flatter, and more crowded, as climate change, pollution, and COVID-19 exacerbate the strain on our environment. The critical question for our generation is this: how do we use the momentum of the first trend to delay or reverse the course of the second?
Revered institutes around the world are focusing on this million-dollar question. One such hallowed hall is Spain’s Institute for Cross-Disciplinary Physics and Complex Systems (Instituto de Fisica Interdisciplinar y Sistemas Complejos; abbreviated as IFISC), a joint research facility founded by the University of the Balearic Islands (UIB) and the Spanish National Research Council (CSIC). Located on the island of Mallorca in the Mediterranean Sea, the IFISC is engaged in interdisciplinary research in various fields of study, including quantum technologies, photonics, environmental sciences, biosystems engineering, and sociotechnical systems.
The IFISC is devoting considerable resources to the most pressing issues. Regarding the environment, there is the “SuMaECO” project, which studies the impact of climate change on aquatic plants in the Mediterranean Sea; and the “Xylella” project, which utilizes machine learning to detect Xylella fastidiosa (Xf), a devastating plant pathogen that’s spreading rapidly, especially during the hotter summers caused by global warming. With regard to pollution and renewable energy, the “ARCTIC” project uses big data to improve air travel efficiency, while the EU-funded “VPP4ISLANDS” project creates digital twins of real locations to optimize the deployment of green energy. Last but not least, “DISTANCIA-COVID” analyzes mobile phone data to understand population movement as it relates to COVID-19.
All of these projects run into a common hurdle. Incredible computing power is needed to carry out high performance computing (HPC), numerical simulations, artificial intelligence (AI) development, and big data management. Not just any servers can do the job. The IFISC needs servers that can satisfy the full gamut of computational needs, because the issues faced by humanity are multifaceted, and there are different approaches to solving the world’s problems.
SuMaECO and Xylella: Protecting the Mediterranean’s Surf and Turf
The “SuMaECO” project focuses on Posidonia oceanica, a seagrass species known as “the lungs of the Mediterranean” due to its ability to absorb carbon dioxide and foster a diverse biosphere. These plants take a century to grow, but satellite mapping and sonar imaging have shown holes in the underwater meadows. The culprit: pollution and global warming. The loss of Posidonia may be a harbinger of greater destruction to the marine environment, which will severely impact the Mediterranean countries. The IFISC wants to use computer simulations to answer a number of questions, such as: how do Posidonia grow? Why are there holes? What must be done to stop the deterioration?
Concurrently, the IFISC is helming the “Xylella” project, which uses machine learning to remotely detect Xylella fastidiosa, a pathogen afflicting 500 plant species around the globe. Xf originated from warmer climes, but has begun to thrive in previously unaffected regions as the world grows hotter. In 2013, it notoriously ravaged the Apulian olive groves in southern Italy. It has since spread as far as Iberia and Israel.
As the world gets warmer, the plant pathogen Xylella fastidiosa is spreading to previously unaffected locations, such as the olive groves of Italy. The IFISC wants to develop an algorithm that can detect possible outbreaks through satellite imagery.
The IFISC wants to detect Xf more quickly and cost-effectively. They are developing an AI to identify the symptoms of Xf via satellite imagery. The process works like this: in each pixel of a picture of a tree, there are dozens of features that may indicate the presence of Xf. An algorithm can be trained through deep learning to calculate the probability of infection, and reach a verdict on the tree’s wellbeing.
Parallel computing is vital for both of these projects. In the case of “SuMaECO”, highly detailed models are used to simulate the growth of Posidonia under different circumstances. In the case of “Xylella”, the algorithm must sift through a massive library of images before it can learn how to differentiate between healthy and stricken plants.《Glossary: What is Parallel Computing?》
Herein lies the IFISC’s first computational need: in order to advance the two equally important projects, they need servers that can boost the processing power of CPUs with GPU accelerators to take care of multiple tasks simultaneously.
ARCTIC and VPP4ISLANDS: Going Green in a Smarter Way
Whereas the first two projects deal directly with nature, “ARCTIC” and “VPP4ISLANDS” dissect the human factor. Both of these projects look for ways to advance technology without aggravating the environmental cost.
“ARCTIC” seeks to solve a common phenomenon in air travel known as “flight delay propagation”. This is what happens when the minuscule delay of one flight cascades into a snowball effect of unplanned layovers, passenger congestion, and wasted time and resources. The research team proposes to use data analytics and machine learning to study air traffic patterns. They want to weed out problematic procedures that cause delays, so they can establish a set of best practices to limit the occurrence of flight delay propagation. This will help to reduce fuel consumption and air pollution.
“VPP4ISLANDS” uses computer models to create “digital twins” of real-world power grids. Scientists use these virtual representations to simulate the implementation of renewable energy in isolated instances—the so-called “islands” in the project name. By doing this, researchers will have the freedom to experiment with disruptive solutions and accelerate the transition towards green energy, without inadvertently causing detrimental consequences. Methods that work well in simulations would be introduced to the real world; they would then be widely implemented, if real-life trials also prove to be successful.
“Flight delay propagation” refers to the snowball effect of wasted time and resources caused by a single delayed flight. The IFISC aims to establish a set of best practices and improve air travel efficiency with data analytics and machine learning.
Both of these projects rely on a combination of innovative technological breakthroughs, such as AI, virtual simulations, data analytics, and HPC. Processing speed is not the only criterion; versatility, multifunctionality, and scalability must also be taken into account.《Glossary: What is Scalability?》
Herein lies the IFISC’s second computational need: in order to juggle the variable requirements of multiple ongoing projects, and in order to prepare for the unknowable requirements of future projects, they need servers that are versatile, accessible, and suitable for all sorts of different tasks.
DISTANCIA-COVID: Vivisecting the Virus by Monitoring Mobile Data
Needless to say, the world faces an additional wrinkle on top of climate change and pollution: COVID-19. Like many academic institutes around the world, the IFISC is helping to study the virus.
The IFISC is contributing through the “DISTANCIA-COVID” project. It analyzes mobile phone data to understand the change in user mobility and its effect on the spread of COVID. It also studies the benefits of social distancing measures. This project calls for massive loads of data to be transferred quickly and stored safely, so herein lies the IFISC’s third computational need: stable storage and low-latency data transfer.
Through Sistemas Informáticos Europeos (SIE), a Spanish company specializing in servers, workstations, and other network and communications systems, the IFISC chose GIGABYTE’s advanced server solutions as the best answer to its computational needs. The IFISC built computing clusters with three types of GIGABYTE servers: the 2U 24-Bay R282-Z91, the 2U 26-bay R272-Z32 (both of which are R-Series Rack Servers), and the 4U 8-node G482-Z54 G-Series GPU Server.
“We are very familiar with the GIGABYTE brand; in fact, many of our office PCs use GIGABYTE motherboards,” says Professor Emilio Hernández-García, Director of the IFISC. “Imagine our pleasant surprise when we learned GIGABYTE also offers server solutions—exactly what we need to take our research to the next level.”
The three types of servers are assigned different roles. The G482-Z54 is the parallel computing powerhouse, well suited for the simulation requirements of the “SuMaECO” project, and the AI training needs of the “Xylella” project. The R282-Z91 is the workhorse model that lends its impressive computing power to just about everything. And the R272-Z32 is the database that excels at big data management, which is especially crucial for the “DISTANCIA-COVID” project.
“GIGABYTE is the main provider of all the IFISC’s computing power. Their servers are behind all the projects involving intensive computing, data analysis, and GPU-related computing,” says Dr. Pere Colet, Computer Systems Scientific Supervisor at the IFISC. “Out of nearly a dozen brands we’ve evaluated, GIGABYTE servers offered us by far the best performance.”
Computational Need #1: AMD CPUs + High-density GPU Accelerators
One thing that’s undeniable about GIGABYTE servers is their prodigious processing power. All three servers are outfitted with AMD EPYC™ 7002 processors, which can house up to 64 cores and 128 threads in a single CPU.
The G482-Z54 is a 4U 8-Node G-Series GPU Server that’s ideal for HPC and parallel computing. The ultra-dense configuration of GPU cards boosts the processing power of the dual AMD EPYC™ processors to new frontiers.
In particular, the G482-Z54 can bolster the speed of the CPUs with a dense array of GPU accelerators, which is a key feature in GIGABYTE’s GPU Servers. The G482-Z54 has a 4U chassis that can house up to 8 GPU cards. These cards support PCIe Gen 4.0, which has a maximum bandwidth of 64GB/s and is twice as fast as PCIe Gen 3.0. The G482-Z54 is a natural choice for parallel computing, high performance computing (HPC), cloud computing, and many other data-intensive applications.
For its G482-Z54, the IFISC chose dual AMD EPYC™ 7282 processors, which has 16 cores and 32 threads in each CPU. They supplemented the powerful processors with 6 NVIDIA GeForce RTX™ GPU cards, also produced by GIGABYTE. This set-up is exactly what the “SuMaECO” and “Xylella” projects needed. Now, the IFISC can precisely simulate the growth of the Mediterranean’s Posidonia meadows, and AI models can also be used to detect signs of Xylella fastidiosa in satellite photos.
Computational Need #2: Versatile, Accessible, and Suitable for Every Task
While the G482-Z54 is a specialist, there are other projects with a wider range of general requirements, including data analytics and virtual simulations. The IFISC needs to augment their data center with an array of versatile, multifunctional servers, capable of handling just about any assignment in a cost-effective manner.《Glossary: What is Data Center?》
This is where the R282-Z91 comes in. Like other models in GIGABYTE’s R-Series of Rack Servers, the R282-Z91 is designed with versatility, efficiency, and reliability in mind. The Rack Servers especially shine when used in conjunction with GIGABYTE’s G-Series GPU Servers and H-Series High Density Servers, because they can help clients achieve performance and budget goals.
This is exactly how the IFISC is using their 20 units of R282-Z91. While the G482-Z54 fills the role of the parallel computing powerhouse, the R282-Z91 is the workhorse model that handles a lot of the heavy lifting. Workloads from a number of projects are distributed across both types of servers, including “ARCTIC”, which seeks to solve flight delay propagation, and “VPP4ISLANDS”, which experiments with green energy using digital twins.
The R282-Z91 features 24 bays in a 2U chassis. The IFISC has a computing cluster made up of 20 nodes of these servers, and they support just about any project that requires high performance computing or intensive numerical simulations.
In addition, the IFISC needs to ensure its servers are easily accessible to different teams working on different projects around the institute. This is why they purchased GIGABYTE’s CLN4C44 alongside the servers. The CLN4C44 is a quad-port LAN add-on card featuring dual Mellanox ConnectX-4 Lx EN network controllers and 25Gb/s Ethernet connectivity. This innovative server accessory solves the issue of accessibility by providing true hardware-based I/O virtualization support with unmatched scalability. Its advanced NVGRE, VXLAN and GENEVE hardware offloading engines can alleviate the high workload placed on CPUs, which results from data center operators using overlay networks to handle traffic from virtual machines in order to better scale their networks. The CLN4C44 is a cost-effective and flexible solution, well suited for hyperscale computing, cloud infrastructures, enterprise data centers, and more.
Computational Need #3: Stable Storage and Low-latency Data Transfer
It goes without saying that the IFISC’s numerous projects generate an enormous amount of data. Especially in the case of the “DISTANCIA-COVID” project, which evaluates COVID-19 countermeasures by studying mobile phone data, the IFISC requires servers that can handle big data management, in addition to working with the cross-platform NoSQL database program MongoDB.
This is why the IFISC chose the R272-Z32, outfitted with a single AMD EPYC™ 7302P processor and 12 solid state drives (SSD) in its 2U chassis, for this special mission. The R272-Z32 fulfills the IFISC’s third computational requirement: stable storage and low-latency data transfer. Its terabytes of storage can be conveniently retrieved via Non-Volatile Memory Express (NVMe), an interface standard that benefits from the quick access and low latency of SSDs, allowing for blazing fast communication between disks.《Glossary: What is NVMe?》
It is also worth noting that GIGABYTE servers are the epitomes of stability. As with the majority of GIGABYTE’s air-cooled servers, the G482-Z54, R282-Z91, and R272-Z32 come with dynamic fan speed control as standard. This is an intelligent system that monitors the temperatures of key components through the baseboard management controller (BMC). Fan speed is automatically adjusted to control the temperature while delivering superb power usage efficiency (PUE).
What’s more, GIGABYTE’s proprietary SCMP (Smart Crises Management/ Protection) feature can force the CPU to enter ultra-low frequency mode (ULFM) if the BMC detects a dangerous fault or error, such as overheating. This safety function prevents the system from shutting down. Once the issue has been resolved, the system will automatically return to normal power mode.
The R272-Z32 is a 2U 26-Bay R-Series Rack Server that’s recommended for networking and storage. It is a versatile machine that is noted for its easy-to-use tool-less design and unparalleled stability.
“At the end of the day, most of us here are scientists, not computer engineers,” says Dr. Pere Colet. “Fortunately, GIGABYTE and SIE were able to understand our specific computational requirements, and offer us the best combination of GIGABYTE servers to solve our problems. We are very happy to be working with them.”
The ongoing effort to solve the world’s problems is a long struggle; it will certainly extend beyond our lifetime. Thankfully, with the world’s brightest minds using the world’s foremost server solutions to tackle these problems, there is reason to be optimistic. GIGABYTE is glad to support the IFISC’s research projects with groundbreaking server solutions. The GIGABYTE motto is “Upgrade Your Life”. It is a sincere belief that cutting-edge technology can be used to advance human prosperity; that tomorrow will be a better place, because we are all in this, together.