Date: April 26, 2023 | 3 p.m. (UTC+1)

The post Webinar: Improving energy-efficiency of High-Performance Computing clusters first appeared on RISC2 Project.

In each webinar, it will be presented the state-of-the-art in methods and tools for setting-up and maintaining HPC hardware and software infrastructures. The duration of each talk will be around 30-40 minutes, followed by a 10–15-minute moderated discussion with the audience.

There are already 4 webinars scheduled:

• Developing complex workflows that include HPC, Artificial Intelligence and Data Analytics | February 22

• A roadmap to quantum computing integration into HPC infrastructures | March 15

• Improving energy-efficiency of High-Performance Computing clusters | April 26

• Addressing the challenges of scientific visualization in the exascale age | May 31

The post RISC2 webinar series aims to benefit HPC research and industry in Europe and Latin America first appeared on RISC2 Project.

Date: February 22, 2023 | 4 p.m. (UTC)

The post Webinar: Developing complex workflows that include HPC, Artificial Intelligence and Data Analytics first appeared on RISC2 Project.

However, to realize the full potential of this synergy, ML models (or models for short) must be built, combined and ensembled, which can be very complex as there can be many models to select from. Furthermore, they should be shared and reused, in particular, in different execution environments such as HPC or Spark clusters.

To address this problem, we proposed Gypscie [Porto 2022, Zorrilla 2022], a new framework that supports the entire ML lifecycle and enables model reuse and import from other frameworks. The approach behind Gypscie is to combine several rich capabilities for model and data management, and model execution, which are typically provided by different tools, in a unique framework. Overall, Gypscie provides: a platform for supporting the complete model life-cycle, from model building to deployment, monitoring and policies enforcement; an environment for casual users to find ready-to-use models that best fit a particular prediction problem, an environment to optimize ML task scheduling and execution; an easy way for developers to benchmark their models against other competitive models and improve them; a central point of access to assess models’ compliance to policies and ethics and obtain and curate observational and predictive data; provenance information and model explainability. Finally, Gypscie interfaces with multiple execution environments to run ML tasks, e.g., an HPC system such as the Santos Dumont supercomputer at LNCC or a Spark cluster. 

Gypscie comes with SAVIME [Silva 2020], a multidimensional array in-memory database system for importing, storing and querying model (tensor) data. The SAVIME open-source system has been developed to support analytical queries over scientific data. Its offers an extremely efficient ingestion procedure, which practically eliminates the waiting time to analyze incoming data. It also supports dense and sparse arrays and non-integer dimension indexing. It offers a functional query language processed by a query optimiser that generates efficient query execution plans.

References

[Porto 2022] Fabio Porto, Patrick Valduriez: Data and Machine Learning Model Management with Gypscie. CARLA 2022 – Workshop on HPC and Data Sciences meet Scientific Computing, SCALAC, Sep 2022, Porto Alegre, Brazil. pp.1-2. 

[Zorrilla 2022] Rocío Zorrilla, Eduardo Ogasawara, Patrick Valduriez, Fabio Porto: A Data-Driven Model Selection Approach to Spatio-Temporal Prediction. SBBD 2022 – Brazilian Symposium on Databases, SBBD, Sep 2022, Buzios, Brazil. pp.1-12. 

[Silva 2020] A.C. Silva, H. Lourenço, D. Ramos, F. Porto, P. Valduriez. Savime: An Array DBMS for Simulation Analysis and Prediction. Journal of Information Data Management 11(3), 2020. 

By LNCC and Inria 

The post Managing Data and Machine Learning Models in HPC Applications first appeared on RISC2 Project.

In Life Sciences, HPC advancements can improve different areas:

• a reduced time to scientific discovery;
• the ability of generating predictions necessary for precision medicine;
• new healthcare and genomics-driven research approaches;
• the processing of huge datasets for deep and machine learning;
• the optimization of modeling, such as Computer Aided Drug Design (CADD);
• enhanched security and protection of healthcare data in HPC environments, in compliance with European GDPR regulations;
• management of massive amount of data for example for clinical trials, drug development and genomics data analytics.

The outbreak of COVID-19 has further accelerated this progress from different points of view. Some European projects aim at reusing known and active ingredients to prepare new drugs as contrast therapy against COVID disease [Exscalate4CoV, Ligate], while others focus on the management and monitoring of contagion clusters to provide an innovative approach to learn from SARS-CoV-2 crisis and derive recommendations for future waves and pandemics [Orchestra].

The ability to deal with massive amounts of data in HPC environments is also used to create databases with data from nucleic acids sequencing and use them to detect allelic variant frequencies, as in the NIG project [Nig], a collaboration with the Network for Italian Genomes. Another example of usage of this capability is the set-up of data sharing platform based on novel Federated Learning schemes, to advance research in personalised medicine in haematological diseases [Genomed4All].

Supercomputing is widely used in Drug Design (the process of finding medicines for disease for which there are no or insufficient treatments), with many projects active in this field just like RISC2.

Sometimes, when there is no previous knowledge of the biological target, just like what happened with COVID-19, discovering new drugs requires creating from scratch new molecules [Novartis]. This process involves billion dollar investments to produce and test thousands of molecules and it usually has a low success rate: only about 12% of potential drugs entering the clinical development are approved [Engitix]. The whole process from identifying a possible compound to the end of the clinical trial can take up to 10 years. Nowadays there is an uneven coverage of disease: most of the compounds are used for genetic conditions, while only a few antiviral and antibiotics have been found.

The search for candidate drugs occurs mainly through two different approaches: high-throughput screening and virtual screening. The first one is more reliable but also very expensive and time consuming: it is usually applied when dealing with well-known targets by mainly pharmaceutical companies. The second approach is a good compromise between cost and accuracy and is typically applied against relatively new targets, in academics laboratories, where it is also used to discover or understand better mechanisms of these targets. [Liu2016]

Candidate drugs are usually small molecules that bind to a specific protein or part of it, inhibiting the usual activity of the protein itself. For example, binding the correct ligand to a vial enzyme may stop viral infection. In the process of virtual screening million of compounds are screened against the target protein at different levels: the most basic one simply takes into account the shape to correctly fit into the protein, at higher level also other features are considered as specific interactions, protein flexibility, solubility, human tolerance, and so on. A “score” is assigned to each docked ligand: compounds with highest score are further studied. With massively parallel computers, we can rapidly filter extremely large molecule databases (e.g. billions of molecules).

The current computational power of HPC clusters allow us to analyze up to 3 million compounds per second [Exscalate]. Even though vaccines were developed remarkably quickly, effective drug treatments for people already suffering from covid-19 were very fresh at the beginning of the pandemic. At that time, supercomputers around the world were asked to help with drug design, a real-world example of the power of Urgent Computing. CINECA participates in Exscalate4cov [Exscalate4Cov], currently the most advanced center of competence for fighting the coronavirus, combining the most powerful supercomputing resources and Artificial Intelligence with experimental facilities and clinical validation. 

References

[Engitix] https://engitix.com/technology/

[Exscalate] https://www.exscalate.eu/en/projects.html

[Exscalate4CoV] https://www.exscalate4cov.eu/

[Genomed4All] https://genomed4all.eu/

[Ligate] https://www.ligateproject.eu/

[Liu2016] T. Liu, D. Lu, H. Zhang, M. Zheng, H. Yang, Ye. Xu, C. Luo, W. Zhu, K. Yu, and H. Jiang, “Applying high-performance computing in drug discovery and molecular simulation” Natl Sci Rev. 2016 Mar; 3(1): 49–63.

[Nig] http://www.nig.cineca.it/

[Novartis] https://www.novartis.com/stories/art-drug-design-technological-age

[Orchestra] https://orchestra-cohort.eu/

By CINECA

The post Using supercomputing for accelerating life science solutions first appeared on RISC2 Project.

As the utility and usage of HPC infrastructures increases, more computational and storage power is required to efficiently handle the amount of targeted applications. In fact, many HPC centers are now aiming at exascale supercomputers supporting at least one exaFLOPs (1018 operations per second), which represents a thousandfold increase in processing power over the first petascale computer deployed in 2008 [RD+15]. Although this is a necessary requirement for handling the increasing number of HPC applications, there are several outstanding challenges that still need to be tackled so that this extra computational power can be fully leveraged. 

Management of large infrastructures and heterogeneous workloads: By adding more compute and storage nodes, one is also increasing the complexity of the overall HPC distributed infrastructure and making it harder to monitor and manage. This complexity is increased due to the need of supporting highly heterogeneous applications that translate into different workloads with specific data storage and processing needs [ECS+17]. For example, on the one hand, traditional scientific modeling and simulation tasks require large slices of computational time, are CPU-bound, and rely on iterative approaches (parametric/stochastic modeling). On the other hand, data-driven Big Data applications contemplate shorter computational tasks, that are I/O bound and, in some cases, have real-time response requirements (i.e., latency-oriented). Also, many of the applications leverage AI and machine learning tools that require specific hardware (e.g., GPUs) in order to be efficient.

Support for general-purpose analytics: The increased heterogeneity also demands that HPC infrastructures are now able to support general-purpose AI and BigData applications that were not designed explicitly to run on specialised HPC hardware [KWG+13]. Therefore, developers are not required to significantly change their applications so that they can execute efficiently at HPC clusters.

Avoiding the storage bottleneck: By only increasing the computational power and improving the management of HPC infrastructures it may still not be possible to fully harmed the capabilities of these infrastructures. In fact, Big Data and AI applications are data-driven and require efficient data storage and retrieval from HPC clusters. With an increasing number of applications and heterogeneous workloads, the storage systems supporting HPC may easily become a bottleneck [YDI+16, ECS+17]. Indeed, as pointed out by several studies, the storage access time is one of the major bottlenecks limiting the efficiency of current and next-generation HPC infrastructures. 

In order to address these challenges, RISC2 partners are exploring: New monitoring and debugging tools that can aid in the analysis of complex AI and Big Data workloads in order to pinpoint potential performance and efficiency bottlenecks, while helping system administrators and developers on troubleshooting these [ENO+21].

Emerging virtualization technologies, such as containers, that enable users to efficiently deploy and execute traditional AI and BigData applications in an HPC environment, without requiring any changes to their source-code [FMP21].  

The Software-Defined Storage paradigm in order to improve the Quality-of-Service (QoS) for HPC’s storage services when supporting hundreds to thousands of data-intensive AI and Big Data applications [DLC+22, MTH+22].  

To sum up, these three research goals, and respective contributions, will enable the next generation of HPC infrastructures and services that can efficiently meet the demands of Big Data and AI workloads. 

References

[DLC+22] Dantas, M., Leitão, D., Cui, P., Macedo, R., Liu, X., Xu, W., Paulo, J., 2022. Accelerating Deep Learning Training Through Transparent Storage Tiering. IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid)  

[ECS+17] Joseph, E., Conway, S., Sorensen, B., Thorp, M., 2017. Trends in the Worldwide HPC Market (Hyperion Presentation). HPC User Forum at HLRS.  

[FMP21] Faria, A., Macedo, R., Paulo, J., 2021. Pods-as-Volumes: Effortlessly Integrating Storage Systems and Middleware into Kubernetes. Workshop on Container Technologies and Container Clouds (WoC’21). 

[KWG+13] Katal, A., Wazid, M. and Goudar, R.H., 2013. Big data: issues, challenges, tools and good practices. International conference on contemporary computing (IC3). 

[NCR+18] Netto, M.A., Calheiros, R.N., Rodrigues, E.R., Cunha, R.L. and Buyya, R., 2018. HPC cloud for scientific and business applications: Taxonomy, vision, and research challenges. ACM Computing Surveys (CSUR). 

[MTH+22] Macedo, R., Tanimura, Y., Haga, J., Chidambaram, V., Pereira, J., Paulo, J., 2022. PAIO: General, Portable I/O Optimizations With Minor Application Modifications. USENIX Conference on File and Storage Technologies (FAST). 

[OG+15] Osseyran, A. and Giles, M. eds., 2015. Industrial applications of high-performance computing: best global practices. 

[RD+15] Reed, D.A. and Dongarra, J., 2015. Exascale computing and big data. Communications of the ACM. 

[ENO+21] Esteves, T., Neves, F., Oliveira, R., Paulo, J., 2021. CaT: Content-aware Tracing and Analysis for Distributed Systems. ACM/IFIP Middleware conference (Middleware). 

[YDI+16] Yildiz, O., Dorier, M., Ibrahim, S., Ross, R. and Antoniu, G., 2016, May. On the root causes of cross-application I/O interference in HPC storage systems. IEEE International Parallel and Distributed Processing Symposium (IPDPS). 

By INESC TEC

The post HPC meets AI and Big Data first appeared on RISC2 Project.