The IBM Cloud blog has recently published a post by Gil Vernik (IBM Research), Michael Behrendt (Distinguished Engineer, Serverless/FaaS at IBM Research), and Josep Sampé (postdoctoral researcher at Universitat Rovira i Virgili).

The blog post explains how to enjoy the flexibility and cost of IaaS with the convenience of serverless thanks to the Lithops framework and IBM Cloud services like IBM Cloud Code Engine or IBM Cloud Functions.

The CloudButton Big Data challenge consists of developing the entire life cycle of a Big Data application including data acquisition and storage, data preprocessing and indexing, and data queries and visualization. The solution must use the Lithops Toolkit and run in the IBM Cloud.

This Challenge is organized by the H2020 CloudButton project and Universitat Rovira i Virgili, with the collaboration of IBM, RedHat and ATOS. The evaluation committee will include representatives from the four institutions. The CloudButton project is developing novel Cloud technologies aiming to democratize Big Data applications in the Cloud. The CloudButton project has created the Lithops Toolkit that will be used to implement the Big Data challenge.

The Challenge is open to students of Computer Engineering in the last courses and Master students, and they can participate in groups of up to three members. Application is open until April 15th.

For more information and registration forms, please check out challenge.cloudbutton.eu

CloudButton released one of the main software outcomes of the project: the Lithops framework. Lithops aims to massively scale the execution of Python code and its dependencies on serverless computing platforms and to monitor the results. Lithops delivers the user’s code into the serverless platform without requiring knowledge of how functions are invoked and run.

The framework provides an extensible backend architecture (compute, storage) that is designed to work with different Cloud providers and on-premise backends (Knative, OpenWhisk). You can edit programs using Lithops in a laptop and then run the unmodified code at scale in the cloud using any cloud provider (IBM Cloud, AWS, Azure, Google Cloud and Alibaba Aliyun).

Check out the Lithops code on GitHub: https://github.com/lithops-cloud/lithops/.

We are happy to announce that the work of the LSDS group at Imperial College London on Faasm, a high-density stateful serverless runtime based on WebAssembly, was accepted at USENIX Annual Technical Conference (ATC 2020). ATC will take place on July 15–17, 2020, in Boston, MA, USA, and is one of the most prestigious systems venues. Kudos to Simon Shillaker and Peter Pietzuch!

Faasm is a high-performance stateful serverless runtime. The goal of the project is enabling fast, efficient serverless big data.

Faasm provides multi-tenant isolation, but also lets functions share regions of memory. These shared memory regions give low-latency concurrent access to data, supporting high-performance distributed serverless applications.

By running WebAssembly, Faasm combines software fault isolation with standard OS tooling to provide security and resource isolation guarantees at low cost. Functions run side-by-side as threads of a single runtime process, with low overheads and fast boot times.

Check out the Faasm code on GitHub: https://github.com/lsds/Faasm.

Check out the paper preprint on Arxiv: Faasm: Lightweight Isolation for Efficient Stateful Serverless Computing.

CloudButton participates with three different success stories to the second edition of the BDV PPP Best Success Story Contest, organized by the BDVe project and BDVA. The objective of this contest is to promote and give visibility to the results produced by the EU projects, to present them in a way that could be better understood by a wider audience, and to support projects in how to present those results in an entertaining, fun and constructive way.

The CloudButton consortium has submitted 3 success stories to the contest: one for each use case considered in the project:

• Serverless computing to scale up genomic alignment algorithms.
• A serverless approach for fast calculation of irrigation needs.
• Serverless computing to annotate dark molecular matter in spatial metabolomics.

17 representatives of all CloudButton partners attended the 3rd consortium meeting which was held in Barcelona, Spain between 8-10 January 2020.

This meeting was mainly focused on presenting the progress of the project during the first year and boosting collaborations between partners. It was a very productive meeting that paved the way for future achievements.

Daniel Barcelona (CloudLab research group, Universitat Rovira i Virgili) presented a research paper at the International Middleware Conference 2019 that took place from 9th to 13th of December at University of California, Davis (USA). Middleware is a single-track conference that brings together academic and industrial delegates who have an interest in the development, optimisation, evaluation and evolution of middleware.

Daniel presented a joint work between 2 CloudButton partners (URV and IMT): On the FaaS track: Building Stateful Distributed Applications with Serverless Architectures. The paper presents Crucial: a system to program highly-concurrent stateful applications with serverless architectures.

Also, URV team presented another paper at the Workshop on Serverless Computing (WoSC) that was held jointly with the Middleware conference. Dr. Pedro García gave a talk entitled FaaS Orchestration of Parallel Workloads, where he analyzes the limitations of current cloud functions orchestration systems to manage parallel fork-join workflows in a scalable and efficient way.

Dr. Ofer Biran (IBM Research Haifa) gave a talk at the European Big Data Value Forum (EBDVF) that took place from 14th to 16th of October in Helsinki, Finland. EBDVF is the main event of the European Big Data and Data-Driven AI Research and Innovation community.

Ofer Biran's talk, entitled Your easy move to serverless computing and radically simplified data processing created a lot of interests. He presented Cloud Button Toolkit, a framework that is being developed as part of the CloudButton project. Ofer took a deep dive into the challenge of how serverless computing can be easily used for a broad range of scenarios, like high-performance computing (HPC), Monte Carlo simulations, and data preprocessing for AI. He also showed how this toolkit has been used in some of the project use cases like metabolomics use case from EMBL.

Dr. Gil Vernik (IBM Research Haifa) gave a talk at the Strata Data Conference 2019 that took place from 23rd to 26th of September in New York. Put on by Cloudera and O'Reilly, Strata Data focuses on using big data to gain business advantage. The event includes tutorials, training sessions, keynotes, ... so it is an excellent opportunity to see how people are using big data to solve real problems in the real world.

Gil Vernik's talk, entitled "Your easy move to serverless computing and radically simplified data processing", presented Cloud Button Toolkit, a framework that is being developed as part of the CloudButton project. Gil took a deep dive into the challenge of how serverless computing can be easily used for a broad range of scenarios, like high-performance computing (HPC), Monte Carlo simulations, and data preprocessing for AI.

Dr. Josep Sampé (Universitat Rovira i Virgili) presented the CloudButton project at the BigMedilytics Event 2019 that took place on the 4th and 5th of September in Valencia, Spain. Under the title of "Big Data: Fueling the transformation of Europe's Healthcare Sector", the event involved organisations covering the key players in the healthcare sector such as healthcare providers, health technology companies, payers, research institutes and academia from across Europe.

The second meeting of the H2020 CloudButton project, hosted by Imperial College London, will take place in London from Wednesday the 17th until Friday the 19th of July 2019.

The CloudButton project has attended the BDV PPP Summit that was held from 26th to 28th June 2019 at the Latvian National Library in Riga.

About the event

The first day of the BDV PPP Summit 2019 was open to everybody and called the BDV PPP Conference, followed by two days of workshops and the BDV PPP Meetup. Activity Groups, Technical and Steering Committees, as well as the BDVA General Assembly, were on the programme. Keynotes combined with panels and a lot of networking opportunities were an ideal chance for network building and for making the projects known to a broader audience.

24 representatives of all CloudButton partners attended the kick-off meeting which was held in Tarragona, Spain between 22-25 January 2019. This meeting represented the official start of the project.

The kick-off meeting was mainly focused on understanding the project background and stablishing common goals. A road map of tasks for the next six months was agreed. The partners laid the foundation for an efficient cooperation and coordination among each other

The CloudButton project, coordinated by Prof. Pedro García from Universitat Rovira i Virgili, is funded by the European Union's Horizon 2020 research and innovation programme.

All partners will meet in Tarragona, Spain to officially launch the project.

CloudButton, with project number 825184, is funded under the topic ICT-12-2018-2020 "Big Data technologies and extreme-scale analytics" of European Union's Horizon 2020 research and innovation programme. The main objective of the project is to create a Serverless Data Analytics Platform that democratizes big data by overly simplifying the overall life cycle and programming model thanks to serverless technologies. To demonstrate the impact of the project, two settings with large data volumes are targeted: bioinformatics (genomics, metabolomics) and geospatial data (LiDAR, satellital).

During the meeting all issues related to management, finance and collaboration of the CloudButton project will be presented and discussed. The first day will be devoted to a plenary session to present an overall view of the project. The next days will be structured in parallel sessions where the participants will focus on the specification of use cases and the technologies to be used.