Data
-
Unraveling Mitochondrial DNA Mutations in Human Cells
Throughout our lifetime, cells accumulate DNA mutations, which contribute to genetic diversity, or “mosaicism”, among cells. These genomic mutations are pivotal for the aging process and the onset of various diseases, including cancer. Mitochondria, essential cellular organelles involved in energy metabolism and apoptosis, possess their own DNA, which are susceptible to mutations. However, studies on mtDNA mutations and mosaicism have been limited due to a variety of technical challenges.
Genomic scientists from KAIST have revealed the genetic mosaicism characterized by variations in mitochondrial DNA (mtDNA) across normal human cells. This study provides fundamental insights into understanding human aging and disease onset mechanisms.
The study, “Mitochondrial DNA mosaicism in normal human somatic cells,” was published in Nature Genetics on July 22. It was led by graduate student Jisong An under the supervision of Professor Young Seok Ju from the Graduate School of Medical Science and Engineering.
Researchers from Seoul National University College of Medicine, Yonsei University College of Medicine, Korea University College of Medicine, Washington University School of Medicine National Cancer Center, Seoul National University Hospital, Gangnam Severance Hospital and KAIST faculty startup company Inocras Inc. also participated in this study.
< Figure 1. a. Flow of experiment. b. Schematic diagram illustrating the origin and dynamics of mtDNA alterations across a lifetime. >
The study involved a bioinformatic analysis of whole-genome sequences from 2,096 single cells obtained from normal human colorectal epithelial tissue, fibroblasts, and blood collected from 31 individuals. The study highlights an average of three significant mtDNA differences between cells, with approximately 6% of these variations confirmed to be inherited as heteroplasmy from the mother.
Moreover, mutations significantly increased during tumorigenesis, with some mutations contributing to instability in mitochondrial RNA. Based on these findings, the study illustrates a computational model that comprehensively elucidates the evolution of mitochondria from embryonic development to aging and tumorigenesis.
This study systematically reveals the mechanisms behind mitochondrial DNA mosaicism in normal human cells, establishing a crucial foundation for understanding the impact of mtDNA on aging and disease onset.
Professor Ju remarked, “By systematically utilizing whole-genome big data, we can illuminate previously unknown phenomena in life sciences.” He emphasized the significance of the study, adding, “For the first time, we have established a method to systematically understand mitochondrial DNA changes occurring during human embryonic development, aging, and cancer development.”
This work was supported by the National Research Foundation of Korea and the Suh Kyungbae Foundation.
2024.07.24 View 1865 -
A KAIST-SNUH Team Devises a Way to Make Mathematical Predictions to find Metabolites Related to Somatic Mutations in Cancers
Cancer is characterized by abnormal metabolic processes different from those of normal cells. Therefore, cancer metabolism has been extensively studied to develop effective diagnosis and treatment strategies. Notable achievements of cancer metabolism studies include the discovery of oncometabolites* and the approval of anticancer drugs by the U.S. Food and Drug Administration (FDA) that target enzymes associated with oncometabolites. Approved anticancer drugs such as ‘Tibsovo (active ingredient: ivosidenib)’ and ‘Idhifa (active ingredient: enasidenib)’ are both used for the treatment of acute myeloid leukemia. Despite such achievements, studying cancer metabolism, especially oncometabolites, remains challenging due to time-consuming and expensive methodologies such as metabolomics. Thus, the number of confirmed oncometabolites is very small although a relatively large number of cancer-associated gene mutations have been well studied.
*Oncometabolite: A metabolite that shows pro-oncogenic function when abnormally accumulated in cancer cells. An oncometabolite is often generated as a result of gene mutations, and this accumulation promotes the growth and survival of cancer cells. Representative oncometabolites include 2-hydroxyglutarate, succinate, and fumarate.
On March 18th, a KAIST research team led by Professor Hyun Uk Kim from the Department of Chemical and Biomolecular Engineering developed a computational workflow that systematically predicts metabolites and metabolic pathways associated with somatic mutations in cancer through collaboration with research teams under Prof Youngil Koh, Prof. Hongseok Yun, and Prof. Chang Wook Jeong from Seoul National University Hospital.
The research teams have successfully reconstructed patient-specific genome-scale metabolic models (GEMs)* for 1,043 cancer patients across 24 cancer types by integrating publicly available cancer patients’ transcriptome data (i.e., from international cancer genome consortiums such as PCAWG and TCGA) into a generic human GEM. The resulting patient-specific GEMs make it possible to predict each patient’s metabolic phenotypes.
*Genome-scale metabolic model (GEM): A computational model that mathematically describes all of the biochemical reactions that take place inside a cell. It allows for the prediction of the cell’s metabolic phenotypes under various genetic and/or environmental conditions.
< Figure 1. Schematic diagram of a computational methodology for predicting metabolites and metabolic pathways associated with cancer somatic mutations. of a computational methodology for predicting metabolites and metabolic pathways associated with cancer somatic mutations. >
The team developed a four-step computational workflow using the patient-specific GEMs from 1,043 cancer patients and somatic mutation data obtained from the corresponding cancer patients. This workflow begins with the calculation of the flux-sum value of each metabolite by simulating the patient-specific GEMs. The flux-sum value quantifies the intracellular importance of a metabolite. Next, the workflow identifies metabolites that appear to be significantly associated with specific gene mutations through a statistical analysis of the predicted flux-sum data and the mutation data. Finally, the workflow selects altered metabolic pathways that significantly contribute to the biosynthesis of the predicted oncometabolite candidates, ultimately generating metabolite-gene-pathway sets as an output.
The two co-first authors, Dr. GaRyoung Lee (currently a postdoctoral fellow at the Dana-Farber Cancer Institute and Harvard Medical School) and Dr. Sang Mi Lee (currently a postdoctoral fellow at Harvard Medical School) said, “The computational workflow developed can systematically predict how genetic mutations affect cellular metabolism through metabolic pathways. Importantly, it can easily be applied to different types of cancer based on the mutation and transcriptome data of cancer patient cohorts.”
Prof. Kim said, “The computational workflow and its resulting prediction outcomes will serve as the groundwork for identifying novel oncometabolites and for facilitating the development of various treatment and diagnosis strategies”.
This study, which was supported by the National Research Foundation of Korea, has been published online in Genome Biology, a representative journal in the field of biotechnology and genetics, under the title "Prediction of metabolites associated with somatic mutations in cancers by using genome‑scale metabolic models and mutation data".
2024.03.18 View 3281 -
NYU-KAIST Global AI & Digital Governance Conference Held
< Photo 1. Opening of NYU-KAIST Global AI & Digital Governance Conference >
In attendance of the Minister of Science and ICT Jong-ho Lee, NYU President Linda G. Mills, and KAIST President Kwang Hyung Lee, KAIST co-hosted the NYU-KAIST Global AI & Digital Governance Conference at the Paulson Center of New York University (NYU) in New York City, USA on September 21st, 9:30 pm.
At the conference, KAIST and NYU discussed the direction and policies for ‘global AI and digital governance’ with participants of upto 300 people which includes scholars, professors, and students involved in the academic field of AI and digitalization from both Korea and the United States and other international backgrounds. This conference was a forum of an international discussion that sought new directions for AI and digital technology take in the future and gathered consensus on regulations.
Following a welcoming address by KAIST President, Kwang Hyung Lee and a congratulatory message from the Minister of Science and ICT, Jong-ho Lee, a panel discussion was held, moderated by Professor Matthew Liao, a graduate of Princeton and Oxford University, currently serving as a professor at NYU and the director at the Center for Bioethics of the NYU School of Global Public Health.
Six prominent scholars took part in the panel discussion. Prof. Kyung-hyun Cho of NYU Applied Mathematics and Data Science Center, a KAIST graduate who has joined the ranks of the world-class in AI language models and Professor Jong Chul Ye, the Director of Promotion Council for Digital Health at KAIST, who is leading innovative research in the field of medical AI working in collaboration with major hospitals at home and abroad was on the panel. Additionally, Professor Luciano Floridi, a founding member of the Yale University Center for Digital Ethics, Professor Shannon Vallor, the Baillie Gifford Professor in the Ethics of Data and Artificial Intelligence at the University of Edinburgh of the UK, Professor Stefaan Verhulst, a Co-Founder and the DIrector of GovLab‘s Data Program at NYU’s Tandon School of Engineering, and Professor Urs Gasser, who is in charge of public policy, governance and innovative technology at the Technical University of Munich, also participated.
Professor Matthew Liao from NYU led the discussion on various topics such as the ways to to regulate AI and digital technologies; the concerns about how deep learning technology being developed in medicinal purposes could be used in warfare; the scope of responsibilities Al scientists' responsibility should carry in ensuring the usage of AI are limited to benign purposes only; the effects of external regulation on the AI model developers and the research they pursue; and on the lessons that can be learned from the regulations in other fields.
During the panel discussion, there was an exchange of ideas about a system of standards that could harmonize digital development and regulatory and social ethics in today’s situation in which digital transformation accelerates technological development at a global level, there is a looming concern that while such advancements are bringing economic vitality it may create digital divides and probles like manipulation of public opinion. Professor Jong-cheol Ye of KAIST (Director of the Promotion Council for Digital Health), in particular, emphasized that it is important to find a point of balance that does not hinder the advancements rather than opting to enforcing strict regulations.
< Photo 2. Panel Discussion in Session at NYU-KAIST Global AI & Digital Governance Conference >
KAIST President Kwang Hyung Lee explained, “At the Digital Governance Forum we had last October, we focused on exploring new governance to solve digital challenges in the time of global digital transition, and this year’s main focus was on regulations.”
“This conference served as an opportunity of immense value as we came to understand that appropriate regulations can be a motivation to spur further developments rather than a hurdle when it comes to technological advancements, and that it is important for us to clearly understand artificial intelligence and consider what should and can be regulated when we are to set regulations on artificial intelligence,” he continued.
Earlier, KAIST signed a cooperation agreement with NYU to build a joint campus, June last year and held a plaque presentation ceremony for the KAIST NYU Joint Campus last September to promote joint research between the two universities. KAIST is currently conducting joint research with NYU in nine fields, including AI and digital research. The KAIST-NYU Joint Campus was conceived with the goal of building an innovative sandbox campus centering aroung science, technology, engineering, and mathematics (STEM) combining NYU's excellent humanities and arts as well as basic science and convergence research capabilities with KAIST's science and technology.
KAIST has contributed to the development of Korea's industry and economy through technological innovation aiding in the nation’s transformation into an innovative nation with scientific and technological prowess. KAIST will now pursue an anchor/base strategy to raise KAIST's awareness in New York through the NYU Joint Campus by establishing a KAIST campus within the campus of NYU, the heart of New York.
2023.09.22 View 5722 -
KAIST researchers find sleep delays more prevalent in countries of particular culture than others
Sleep has a huge impact on health, well-being and productivity, but how long and how well people sleep these days has not been accurately reported. Previous research on how much and how well we sleep has mostly relied on self-reports or was confined within the data from the unnatural environments of the sleep laboratories.
So, the questions remained: Is the amount and quality of sleep purely a personal choice? Could they be independent from social factors such as culture and geography?
< From left to right, Sungkyu Park of Kangwon National University, South Korea; Assem Zhunis of KAIST and IBS, South Korea; Marios Constantinides of Nokia Bell Labs, UK; Luca Maria Aiello of the IT University of Copenhagen, Denmark; Daniele Quercia of Nokia Bell Labs and King's College London, UK; and Meeyoung Cha of IBS and KAIST, South Korea >
A new study led by researchers at Korea Advanced Institute of Science and Technology (KAIST) and Nokia Bell Labs in the United Kingdom investigated the cultural and individual factors that influence sleep. In contrast to previous studies that relied on surveys or controlled experiments at labs, the team used commercially available smartwatches for extensive data collection, analyzing 52 million logs collected over a four-year period from 30,082 individuals in 11 countries. These people wore Nokia smartwatches, which allowed the team to investigate country-specific sleep patterns based on the digital logs from the devices.
< Figure comparing survey and smartwatch logs on average sleep-time, wake-time, and sleep durations. Digital logs consistently recorded delayed hours of wake- and sleep-time, resulting in shorter sleep durations. >
Digital logs collected from the smartwatches revealed discrepancies in wake-up times and sleep-times, sometimes by tens of minutes to an hour, from the data previously collected from self-report assessments. The average sleep-time overall was calculated to be around midnight, and the average wake-up time was 7:42 AM. The team discovered, however, that individuals' sleep is heavily linked to their geographical location and cultural factors. While wake-up times were similar, sleep-time varied by country. Individuals in higher GDP countries had more records of delayed bedtime. Those in collectivist culture, compared to individualist culture, also showed more records of delayed bedtime. Among the studied countries, Japan had the shortest total sleep duration, averaging a duration of under 7 hours, while Finland had the longest, averaging 8 hours.
Researchers calculated essential sleep metrics used in clinical studies, such as sleep efficiency, sleep duration, and overslept hours on weekends, to analyze the extensive sleep patterns. Using Principal Component Analysis (PCA), they further condensed these metrics into two major sleep dimensions representing sleep quality and quantity. A cross-country comparison revealed that societal factors account for 55% of the variation in sleep quality and 63% of the variation in sleep quantity.
Countries with a higher individualism index (IDV), which placed greater emphasis on individual achievements and relationships, had significantly longer sleep durations, which could be attributed to such societies having a norm of going to bed early. Spain and Japan, on the other hand, had the bedtime scheduled at the latest hours despite having the highest collectivism scores (low IDV). The study also discovered a moderate relationship between a higher uncertainty avoidance index (UAI), which measures implementation of general laws and regulation in daily lives of regular citizens, and better sleep quality.
Researchers also investigated how physical activity can affect sleep quantity and quality to see if individuals can counterbalance cultural influences through personal interventions. They discovered that increasing daily activity can improve sleep quality in terms of shortened time needed in falling asleep and waking up. Individuals who exercise more, however, did not sleep longer. The effect of exercise differed by country, with more pronounced effects observed in some countries, such as the United States and Finland. Interestingly, in Japan, no obvious effect of exercise could be observed. These findings suggest that the relationship between daily activity and sleep may differ by country and that different exercise regimens may be more effective in different cultures.
This research published on the Scientific Reports by the international journal, Nature, sheds light on the influence of social factors on sleep. (Paper Title "Social dimensions impact individual sleep quantity and quality" Article number: 9681)
One of the co-authors, Daniele Quercia, commented: “Excessive work schedules, long working hours, and late bedtime in high-income countries and social engagement due to high collectivism may cause bedtimes to be delayed.”
Commenting on the research, the first author Shaun Sungkyu Park said, "While it is intriguing to see that a society can play a role in determining the quantity and quality of an individual's sleep with large-scale data, the significance of this study is that it quantitatively shows that even within the same culture (country), individual efforts such as daily exercise can have a positive impact on sleep quantity and quality."
"Sleep not only has a great impact on one’s well-being but it is also known to be associated with health issues such as obesity and dementia," said the lead author, Meeyoung Cha. "In order to ensure adequate sleep and improve sleep quality in an aging society, not only individual efforts but also a social support must be provided to work together," she said. The research team will contribute to the development of the high-tech sleep industry by making a code that easily calculates the sleep indicators developed in this study available free of charge, as well as providing the benchmark data for various types of sleep research to follow.
2023.07.07 View 4882 -
A KAIST research team develops a high-performance modular SSD system semiconductor
In recent years, there has been a rise in demand for large amounts of data to train AI models and, thus, data size has become increasingly important over time. Accordingly, solid state drives (SSDs, storage devices that use a semiconductor memory unit), which are core storage devices for data centers and cloud services, have also seen an increase in demand. However, the internal components of higher performing SSDs have become more tightly coupled, and this tightly-coupled structure limits SSD from maximized performance.
On June 15, a KAIST research team led by Professor Dongjun Kim (John Kim) from the School of Electrical Engineering (EE) announced the development of the first SSD system semiconductor structure that can increase the reading/writing performance of next generation SSDs and extend their lifespan through high-performance modular SSD systems.
Professor Kim’s team identified the limitations of the tightly-coupled structures in existing SSD designs and proposed a de-coupled structure that can maximize SSD performance by configuring an internal on-chip network specialized for flash memory. This technique utilizes on-chip network technology, which can freely send packet-based data within the chip and is often used to design non-memory system semiconductors like CPUs and GPUs. Through this, the team developed a ‘modular SSD’, which shows reduced interdependence between front-end and back-end designs, and allows their independent design and assembly.
*on-chip network: a packet-based connection structure for the internal components of system semiconductors like CPUs/GPUs. On-chip networks are one of the most critical design components for high-performing system semiconductors, and their importance grows with the size of the semiconductor chip.
Professor Kim’s team refers to the components nearer to the CPU as the front-end and the parts closer to the flash memory as back-end. They newly constructed an on-chip network specific to flash memory in order to allow data transmission between the back-end’s flash controller, proposing a de-coupled structure that can minimize performance drop.
The SSD can accelerate some functions of the flash translation layer, a critical element to drive the SSD, in order to allow flash memory to actively overcome its limitations. Another advantage of the de-coupled, modular structure is that the flash translation layer is not limited to the characteristics of specific flash memories. Instead, their front-end and back-end designs can be carried out independently. Through this, the team could produce 21-times faster response times compared to existing systems and extend SSD lifespan by 23% by also applying the DDS defect detection technique.
< Figure 1. Schematic diagram of the structure of a high-performance modular SSD system developed by Professor Dong-Jun Kim's team >
This research, conducted by first author and Ph.D. candidate Jiho Kim from the KAIST School of EE and co-author Professor Myoungsoo Jung, was presented on the 19th of June at the 50th IEEE/ACM International Symposium on Computer Architecture, the most prestigious academic conference in the field of computer architecture, held in Orlando, Florida. (Paper Title: Decoupled SSD: Rethinking SSD Architecture through Network-based Flash Controllers)
< Figure 2. Conceptual diagram of hardware acceleration through high-performance modular SSD system >
Professor Dongjun Kim, who led the research, said, “This research is significant in that it identified the structural limitations of existing SSDs, and showed that on-chip network technology based on system memory semiconductors like CPUs can drive the hardware to actively carry out the necessary actions. We expect this to contribute greatly to the next-generation high-performance SSD market.” He added, “The de-coupled architecture is a structure that can actively operate to extend devices’ lifespan. In other words, its significance is not limited to the level of performance and can, therefore, be used for various applications.”
KAIST commented that this research is also meaningful in that the results were reaped through a collaborative study between two world-renowned researchers: Professor Myeongsoo Jung, recognized in the field of computer system storage devices, and Professor Dongjun Kim, a leading researcher in computer architecture and interconnection networks.
This research was funded by the National Research Foundation of Korea, Samsung Electronics, the IC Design Education Center, and Next Generation Semiconductor Technology and Development granted by the Institute of Information & Communications Technology, Planning & Evaluation.
2023.06.23 View 4547 -
Yuji Roh Awarded 2022 Microsoft Research PhD Fellowship
KAIST PhD candidate Yuji Roh of the School of Electrical Engineering (advisor: Prof. Steven Euijong Whang) was selected as a recipient of the 2022 Microsoft Research PhD Fellowship.
< KAIST PhD candidate Yuji Roh (advisor: Prof. Steven Euijong Whang) >
The Microsoft Research PhD Fellowship is a scholarship program that recognizes outstanding graduate students for their exceptional and innovative research in areas relevant to computer science and related fields. This year, 36 people from around the world received the fellowship, and Yuji Roh from KAIST EE is the only recipient from universities in Korea. Each selected fellow will receive a $10,000 scholarship and an opportunity to intern at Microsoft under the guidance of an experienced researcher.
Yuji Roh was named a fellow in the field of “Machine Learning” for her outstanding achievements in Trustworthy AI. Her research highlights include designing a state-of-the-art fair training framework using batch selection and developing novel algorithms for both fair and robust training. Her works have been presented at the top machine learning conferences ICML, ICLR, and NeurIPS among others. She also co-presented a tutorial on Trustworthy AI at the top data mining conference ACM SIGKDD. She is currently interning at the NVIDIA Research AI Algorithms Group developing large-scale real-world fair AI frameworks.
The list of fellowship recipients and the interview videos are displayed on the Microsoft webpage and Youtube.
The list of recipients: https://www.microsoft.com/en-us/research/academic-program/phd-fellowship/2022-recipients/
Interview (Global): https://www.youtube.com/watch?v=T4Q-XwOOoJc
Interview (Asia): https://www.youtube.com/watch?v=qwq3R1XU8UE
[Highlighted research achievements by Yuji Roh: Fair batch selection framework]
[Highlighted research achievements by Yuji Roh: Fair and robust training framework]
2022.10.28 View 8613 -
Machine Learning-Based Algorithm to Speed up DNA Sequencing
The algorithm presents the first full-fledged, short-read alignment software that leverages learned indices for solving the exact match search problem for efficient seeding
The human genome consists of a complete set of DNA, which is about 6.4 billion letters long. Because of its size, reading the whole genome sequence at once is challenging. So scientists use DNA sequencers to produce hundreds of millions of DNA sequence fragments, or short reads, up to 300 letters long. Then the DNA sequencer assembles all the short reads like a giant jigsaw puzzle to reconstruct the entire genome sequence. Even with very fast computers, this job can take hours to complete.
A research team at KAIST has achieved up to 3.45x faster speeds by developing the first short-read alignment software that uses a recent advance in machine-learning called a learned index.
The research team reported their findings on March 7, 2022 in the journal Bioinformatics. The software has been released as open source and can be found on github (https://github.com/kaist-ina/BWA-MEME).
Next-generation sequencing (NGS) is a state-of-the-art DNA sequencing method. Projects are underway with the goal of producing genome sequencing at population scale. Modern NGS hardware is capable of generating billions of short reads in a single run. Then the short reads have to be aligned with the reference DNA sequence. With large-scale DNA sequencing operations running hundreds of next-generation sequences, the need for an efficient short read alignment tool has become even more critical. Accelerating the DNA sequence alignment would be a step toward achieving the goal of population-scale sequencing. However, existing algorithms are limited in their performance because of their frequent memory accesses.
BWA-MEM2 is a popular short-read alignment software package currently used to sequence the DNA. However, it has its limitations. The state-of-the-art alignment has two phases – seeding and extending. During the seeding phase, searches find exact matches of short reads in the reference DNA sequence. During the extending phase, the short reads from the seeding phase are extended. In the current process, bottlenecks occur in the seeding phase. Finding the exact matches slows the process.
The researchers set out to solve the problem of accelerating the DNA sequence alignment. To speed the process, they applied machine learning techniques to create an algorithmic improvement. Their algorithm, BWA-MEME (BWA-MEM emulated) leverages learned indices to solve the exact match search problem. The original software compared one character at a time for an exact match search. The team’s new algorithm achieves up to 3.45x faster speeds in seeding throughput over BWA-MEM2 by reducing the number of instructions by 4.60x and memory accesses by 8.77x. “Through this study, it has been shown that full genome big data analysis can be performed faster and less costly than conventional methods by applying machine learning technology,” said Professor Dongsu Han from the School of Electrical Engineering at KAIST.
The researchers’ ultimate goal was to develop efficient software that scientists from academia and industry could use on a daily basis for analyzing big data in genomics. “With the recent advances in artificial intelligence and machine learning, we see so many opportunities for designing better software for genomic data analysis. The potential is there for accelerating existing analysis as well as enabling new types of analysis, and our goal is to develop such software,” added Han.
Whole genome sequencing has traditionally been used for discovering genomic mutations and identifying the root causes of diseases, which leads to the discovery and development of new drugs and cures. There could be many potential applications. Whole genome sequencing is used not only for research, but also for clinical purposes. “The science and technology for analyzing genomic data is making rapid progress to make it more accessible for scientists and patients. This will enhance our understanding about diseases and develop a better cure for patients of various diseases.”
The research was funded by the National Research Foundation of the Korean government’s Ministry of Science and ICT.
-PublicationYoungmok Jung, Dongsu Han, “BWA-MEME:BWA-MEM emulated with a machine learning approach,” Bioinformatics, Volume 38, Issue 9, May 2022
(https://doi.org/10.1093/bioinformatics/btac137)
-ProfileProfessor Dongsu HanSchool of Electrical EngineeringKAIST
2022.05.10 View 6963 -
CXL-Based Memory Disaggregation Technology Opens Up a New Direction for Big Data Solution Frameworks
A KAIST team’s compute express link (CXL) provides new insights on memory disaggregation and ensures direct access and high-performance capabilities
A team from the Computer Architecture and Memory Systems Laboratory (CAMEL) at KAIST presented a new compute express link (CXL) solution whose directly accessible, and high-performance memory disaggregation opens new directions for big data memory processing. Professor Myoungsoo Jung said the team’s technology significantly improves performance compared to existing remote direct memory access (RDMA)-based memory disaggregation.
CXL is a peripheral component interconnect-express (PCIe)-based new dynamic multi-protocol made for efficiently utilizing memory devices and accelerators. Many enterprise data centers and memory vendors are paying attention to it as the next-generation multi-protocol for the era of big data.
Emerging big data applications such as machine learning, graph analytics, and in-memory databases require large memory capacities. However, scaling out the memory capacity via a prior memory interface like double data rate (DDR) is limited by the number of the central processing units (CPUs) and memory controllers. Therefore, memory disaggregation, which allows connecting a host to another host’s memory or memory nodes, has appeared.
RDMA is a way that a host can directly access another host’s memory via InfiniBand, the commonly used network protocol in data centers. Nowadays, most existing memory disaggregation technologies employ RDMA to get a large memory capacity. As a result, a host can share another host’s memory by transferring the data between local and remote memory.
Although RDMA-based memory disaggregation provides a large memory capacity to a host, two critical problems exist. First, scaling out the memory still needs an extra CPU to be added. Since passive memory such as dynamic random-access memory (DRAM), cannot operate by itself, it should be controlled by the CPU. Second, redundant data copies and software fabric interventions for RDMA-based memory disaggregation cause longer access latency. For example, remote memory access latency in RDMA-based memory disaggregation is multiple orders of magnitude longer than local memory access.
To address these issues, Professor Jung’s team developed the CXL-based memory disaggregation framework, including CXL-enabled customized CPUs, CXL devices, CXL switches, and CXL-aware operating system modules. The team’s CXL device is a pure passive and directly accessible memory node that contains multiple DRAM dual inline memory modules (DIMMs) and a CXL memory controller. Since the CXL memory controller supports the memory in the CXL device, a host can utilize the memory node without processor or software intervention. The team’s CXL switch enables scaling out a host’s memory capacity by hierarchically connecting multiple CXL devices to the CXL switch allowing more than hundreds of devices. Atop the switches and devices, the team’s CXL-enabled operating system removes redundant data copy and protocol conversion exhibited by conventional RDMA, which can significantly decrease access latency to the memory nodes.
In a test comparing loading 64B (cacheline) data from memory pooling devices, CXL-based memory disaggregation showed 8.2 times higher data load performance than RDMA-based memory disaggregation and even similar performance to local DRAM memory. In the team’s evaluations for a big data benchmark such as a machine learning-based test, CXL-based memory disaggregation technology also showed a maximum of 3.7 times higher performance than prior RDMA-based memory disaggregation technologies.
“Escaping from the conventional RDMA-based memory disaggregation, our CXL-based memory disaggregation framework can provide high scalability and performance for diverse datacenters and cloud service infrastructures,” said Professor Jung. He went on to stress, “Our CXL-based memory disaggregation research will bring about a new paradigm for memory solutions that will lead the era of big data.”
-Profile: Professor Myoungsoo Jung Computer Architecture and Memory Systems Laboratory (CAMEL)http://camelab.org School of Electrical EngineeringKAIST
2022.03.16 View 19633 -
‘Game&Art: Auguries of Fantasy’ Features Future of the Metaverse
‘Game & Art: Auguries of Fantasy,’ a special exhibition combining art and technology will feature the new future of metaverse fantasy. The show will be hosted at the Daejeon Creative Center at the Daejeon Museum of Art through September 5.
This show exhibits a combination of science and technology with culture and arts, and introduces young artists whose creativity will lead to new opportunities in games and art. The Graduate School of Culture Technology was designated as a leading culture content academy in 2020 by the Ministry of Culture, Sports & Tourism and the Korea Creative Content Agency for fostering the R&D workforce in creative culture technology.
NCsoft sponsored the show and also participated as an artist. It combined its game-composing elements and technologies with other genres, including data for game construction, scenarios for forming a worldview, and game art and sound. All of the contents can be experienced online in a virtual space as well as offline, and can be easily accessed through personal devices.
Characterized by the themes ‘timeless’ and ‘spaceless’ which connect the past, present, and future, and space created in the digital world. The exhibition gives audience members an opportunity to experience freedom beyond the constraints of time and space under the theme of a fantasy reality created by games and art.
"Computer games, which began in the 1980s, have become cultural content that spans generations, and games are now the fusion field for leading-edge technologies including computer graphics, sound, human-computer interactions, big data, and AI. They are also the best platform for artistic creativity by adding human imagination to technology," said Professor Joo-Han Nam from the Graduate School of Culture Technology, who led the project.
"Our artists wanted to convey various messages to our society through works that connect the past, present, and future through games."
Ju-young Oh's "Unexpected Scenery V2" and "Hope for Rats V2" display game-type media work that raises issues surrounding technology, such as the lack of understanding behind various scientific achievements, the history of accidental achievements, and the side effects of new conveniences.
Tae-Wan Kim, in his work themed ‘healing’ combined the real-time movement of particles which follows the movements of people recorded as digital data. Metadata is collected by sensors in the exhibition space, and floating particle forms are evolved into abstract graphic designs according to audio-visual responses.
Meanwhile, ‘SOS’ is a collaboration work from six KAIST researchers (In-Hwa Yeom, Seung-Eon Lee, Seong-Jin Jeon, Jin-Seok Hong, Hyung-Seok Yoon, and Sang-Min Lee). SOS is based on diverse perspectives embracing phenomena surrounding contemporary natural resources. Audience members follow a gamified path between the various media-elements composing the art’s environment. Through this process, the audience can experience various emotions such as curiosity, suspicion, and recovery.
‘Diversity’ by Sung-Hyun Kim uses devices that recognize the movements of hands and fingers to provide experiences exploring the latent space of game play images learned by deep neural networks. Image volumes generated by neural networks are visualized through physics-based, three-dimensional, volume-rendering algorithms, and a series of processes were implemented based on the self-written code.
2021.06.21 View 7208 -
‘Urban Green Space Affects Citizens’ Happiness’
Study finds the relationship between green space, the economy, and happiness
A recent study revealed that as a city becomes more economically developed, its citizens’ happiness becomes more directly related to the area of urban green space.
A joint research project by Professor Meeyoung Cha of the School of Computing and her collaborators studied the relationship between green space and citizen happiness by analyzing big data from satellite images of 60 different countries.
Urban green space, including parks, gardens, and riversides not only provides aesthetic pleasure, but also positively affects our health by promoting physical activity and social interactions. Most of the previous research attempting to verify the correlation between urban green space and citizen happiness was based on few developed countries. Therefore, it was difficult to identify whether the positive effects of green space are global, or merely phenomena that depended on the economic state of the country. There have also been limitations in data collection, as it is difficult to visit each location or carry out investigations on a large scale based on aerial photographs.
The research team used data collected by Sentinel-2, a high-resolution satellite operated by the European Space Agency (ESA) to investigate 90 green spaces from 60 different countries around the world. The subjects of analysis were cities with the highest population densities (cities that contain at least 10% of the national population), and the images were obtained during the summer of each region for clarity. Images from the northern hemisphere were obtained between June and September of 2018, and those from the southern hemisphere were obtained between December of 2017 and February of 2018.
The areas of urban green space were then quantified and crossed with data from the World Happiness Report and GDP by country reported by the United Nations in 2018. Using these data, the relationships between green space, the economy, and citizen happiness were analyzed.
The results showed that in all cities, citizen happiness was positively correlated with the area of urban green space regardless of the country’s economic state. However, out of the 60 countries studied, the happiness index of the bottom 30 by GDP showed a stronger correlation with economic growth. In countries whose gross national income (GDP per capita) was higher than 38,000 USD, the area of green space acted as a more important factor affecting happiness than economic growth. Data from Seoul was analyzed to represent South Korea, and showed an increased happiness index with increased green areas compared to the past.
The authors point out their work has several policy-level implications. First, public green space should be made accessible to urban dwellers to enhance social support. If public safety in urban parks is not guaranteed, its positive role in social support and happiness may diminish. Also, the meaning of public safety may change; for example, ensuring biological safety will be a priority in keeping urban parks accessible during the COVID-19 pandemic.
Second, urban planning for public green space is needed for both developed and developing countries. As it is challenging or nearly impossible to secure land for green space after the area is developed, urban planning for parks and green space should be considered in developing economies where new cities and suburban areas are rapidly expanding.
Third, recent climate changes can present substantial difficulty in sustaining urban green space. Extreme events such as wildfires, floods, droughts, and cold waves could endanger urban forests while global warming could conversely accelerate tree growth in cities due to the urban heat island effect. Thus, more attention must be paid to predict climate changes and discovering their impact on the maintenance of urban green space.
“There has recently been an increase in the number of studies using big data from satellite images to solve social conundrums,” said Professor Cha. “The tool developed for this investigation can also be used to quantify the area of aquatic environments like lakes and the seaside, and it will now be possible to analyze the relationship between citizen happiness and aquatic environments in future studies,” she added.
Professor Woo Sung Jung from POSTECH and Professor Donghee Wohn from the New Jersey Institute of Technology also joined this research. It was reported in the online issue of EPJ Data Science on May 30.
-PublicationOh-Hyun Kwon, Inho Hong, Jeasurk Yang, Donghee Y. Wohn, Woo-Sung Jung, andMeeyoung Cha, 2021. Urban green space and happiness in developed countries. EPJ Data Science. DOI: https://doi.org/10.1140/epjds/s13688-021-00278-7
-ProfileProfessor Meeyoung ChaData Science Labhttps://ds.ibs.re.kr/
School of Computing
KAIST
2021.06.21 View 9283 -
Advanced NVMe Controller Technology for Next Generation Memory Devices
KAIST researchers advanced non-volatile memory express (NVMe) controller technology for next generation information storage devices, and made this new technology named ‘OpenExpress’ freely available to all universities and research institutes around the world to help reduce the research cost in related fields.
NVMe is a communication protocol made for high-performance storage devices based on a peripheral component interconnect-express (PCI-E) interface. NVMe has been developed to take the place of the Serial AT Attachment (SATA) protocol, which was developed to process data on hard disk drives (HDDs) and did not perform well in solid state drives (SSDs).
Unlike HDDs that use magnetic spinning disks, SSDs use semiconductor memory, allowing the rapid reading and writing of data. SSDs also generate less heat and noise, and are much more compact and lightweight.
Since data processing in SSDs using NVMe is up to six times faster than when SATA is used, NVMe has become the standard protocol for ultra-high speed and volume data processing, and is currently used in many flash-based information storage devices.
Studies on NVMe continue at both the academic and industrial levels, however, its poor accessibility is a drawback. Major information and communications technology (ICT) companies around the world expend astronomical costs to procure intellectual property (IP) related to hardware NVMe controllers, necessary for the use of NVMe. However, such IP is not publicly disclosed, making it difficult to be used by universities and research institutes for research purposes.
Although a small number of U.S. Silicon Valley startups provide parts of their independently developed IP for research, the cost of usage is around 34,000 USD per month. The costs skyrocket even further because each copy of single-use source code purchased for IP modification costs approximately 84,000 USD.
In order to address these issues, a group of researchers led by Professor Myoungsoo Jung from the School of Electrical Engineering at KAIST developed a next generation NVMe controller technology that achieved parallel data input/output processing for SSDs in a fully hardware automated form.
The researchers presented their work at the 2020 USENIX Annual Technical Conference (USENIX ATC ’20) in July, and released it as an open research framework named ‘OpenExpress.’
This NVMe controller technology developed by Professor Jung’s team comprises a wide range of basic hardware IP and key NVMe IP cores. To examine its actual performance, the team made an NVMe hardware controller prototype using OpenExpress, and designed all logics provided by OpenExpress to operate at high frequency.
The field-programmable gate array (FPGA) memory card prototype developed using OpenExpress demonstrated increased input/output data processing capacity per second, supporting up to 7 gigabit per second (GB/s) bandwidth. This makes it suitable for ultra-high speed and volume next generation memory device research.
In a test comparing various storage server loads on devices, the team’s FPGA also showed 76% higher bandwidth and 68% lower input/output delay compared to Intel’s new high performance SSD (Optane SSD), which is sufficient for many researchers studying systems employing future memory devices. Depending on user needs, silicon devices can be synthesized as well, which is expected to further enhance performance.
The NVMe controller technology of Professor Jung’s team can be freely used and modified under the OpenExpress open-source end-user agreement for non-commercial use by all universities and research institutes. This makes it extremely useful for research on next-generation memory compatible NVMe controllers and software stacks.
“With the product of this study being disclosed to the world, universities and research institutes can now use controllers that used to be exclusive for only the world’s biggest companies, at no cost,ˮ said Professor Jung. He went on to stress, “This is a meaningful first step in research of information storage device systems such as high-speed and volume next generation memory.”
This work was supported by a grant from MemRay, a company specializing in next generation memory development and distribution.
More details about the study can be found at http://camelab.org.
Image credit: Professor Myoungsoo Jung, KAIST
Image usage restrictions: News organizations may use or redistribute these figures and image, with proper attribution, as part of news coverage of this paper only.
-Publication:
Myoungsoo Jung. (2020). OpenExpress: Fully Hardware Automated Open Research Framework for Future Fast NVMe Devices. Presented in the Proceedings of the 2020 USENIX Annual Technical Conference (USENIX ATC ’20), Available online at https://www.usenix.org/system/files/atc20-jung.pdf
-Profile: Myoungsoo Jung
Associate Professor
m.jung@kaist.ac.kr
http://camelab.org
Computer Architecture and Memory Systems Laboratory
School of Electrical Engineering
http://kaist.ac.kr
Korea Advanced Institute of Science and Technology (KAIST)
Daejeon, Republic of Korea
(END)
2020.09.04 View 9562 -
Research on the Million Follower Fallacy Receives the Test of Time Award
Professor Meeyoung Cha’s research investigating the correlation between the number of followers on social media and its influence was re-highlighted after 10 years of publication of the paper.
Saying that her research is still as relevant today as the day it was published 10 years ago, the Association for the Advancement of Artificial Intelligence (AAAI) presented Professor Cha from the School of Computing with the Test of Time Award during the 14th International Conference on Web and Social Media (ICWSM) held online June 8 through 11.
In her 2010 paper titled ‘Measuring User Influence in Twitter: The Million Follower Fallacy,’ Professor Cha proved that number of followers does not match the influential power. She investigated the data including 54,981,152 user accounts, 1,963,263,821 social links, and 1,755,925,520 Tweets, collected with 50 servers.
The research compares and illustrates the limitations of various methods used to measure the influence a user has on a social networking platform. These results provided new insights and interpretations to the influencer selection algorithm used to maximize the advertizing impact on big social networking platforms.
The research also looked at how long an influential user was active for, and whether the user could freely cross the borders between fields and be influential on different topics as well. By analyzing cases of who becomes an influencer when new events occur, it was shown that a person could quickly become an influencer using several key tactics, unlike what was previously claimed by the ‘accidental influential theory’.
Professor Cha explained, “At the time, data from social networking platforms did not receive much attention in computer science, but I remember those all-nighters I pulled to work on this project, fascinated by the fact that internet data could be used to solve difficult social science problems. I feel so grateful that my research has been endeared for such a long time.”
Professor Cha received both her undergraduate and graduate degrees from KAIST, and conducted this research during her postdoctoral course at the Max Planck Institute in Germany. She now also serves as a chief investigator of a data science group at the Institute for Basic Science (IBS).
(END)
2020.06.22 View 6226