Archives: Events


London Open Source Meetup for RISC-V

We are resuming our quarterly meetup for the London open source community, focusing on RISC-V, hosted by the BCS Open Source Specialist Group and the UK Open Source Hardware User Group.  These meetings provide an opportunity to share the latest ideas around open source in the RISC-V ecosystem, combined with plenty of time for networking.

This time it will be a virtual meetup, and we’ll be livestreaming using BigBlueButton to provide a rich online experience for participants. As always the talks will be recorded for later upload to YouTube. When you register you’ll also get a link to a GoToWebinar session, which we will use as fallback if there is a problem with BigBlueButton, since we are still learning the technology.

Please register using the link above. For any with connection issues, we’ll also tweet the link on @BCSOSSG just before the meeting starts.

Porting Rust to 64-bit RISC-V GNU+Linux

Tom Eccles

This year several GNU+Linux distributions have released stable 64-bit RISC-V ports; which lead users and package maintainers to expect access to the Rust toolchain on RISC-V. I spent some time getting Rust’s 64-bit RISC-V target to the point that the project felt comfortable publishing official RISC-V builds. This talk explains how to port Rust to a new target and talks through some bugs fixed along the way.

Tom is interested in free software, cryptography and embedded systems. In 2017 he published a paper through MDPI. The paper suggests and evaluates the performance of a privacy preserving smart metering protocol. In 2018 Tom graduated with a masters degree in electronic engineering from the University of Southampton and started work as a software engineer at Codethink. At Codethink Tom helps clients align their embedded GNU+Linux systems more closely with upstream free software projects. Recently, Codethink sponsored Tom to improve support for RISC-V in the Rust ecosystem.

A Plan 9 C Compiler for RV32GC and RV64GC

The Plan 9 C compiler originated at Bell Labs alongside the Plan 9 and Inferno operating systems. It was part of the initial development toolchain for Go (before that language became self-compiling), and is a useful stand alone tool for building embedded software. A new RISC-V version of the Plan 9 C compiler, assembler and linker, targeting RV32IM cores such as PicoRV32, was announced by the author at ORCONF in 2018. The toolchain has now been expanded with floating point, compressed, and 64-bit instruction capability.

Richard Miller is a consulting engineer working in the borderland between software and hardware, on operating systems, programming language implementation, and digital logic design. His first C programming project was in 1977, adapting Dennis Ritchie’s original 16-bit PDP-11 C compiler to generate code for the 32-bit Interdata 7/32.

Porting the GNU CORE-V Toolchain

Mary Bennett
Pietra F T Madio, @7pietraferreira
Jessica Mills

Over recent years, RISC-V has taken off in both academia and industry. One RISC-V variant, CORE-V, developed by the Open Hardware Group, adds extensions which can improve performance and reduce code size. As part of this effort, the Open Hardware group have decided to build a robust set of GNU tools targeting CORE-V.

In the first part of this presentation, we will discuss adding support for hardware loop instructions to GNU binutils. This has involved making changes to the GNU assembler, gas, and the GNU linker, ld. We will explain in technical detail how we ported this custom extension to binutils as well how we tested the resulting CORE-V assembler and linker.

The second part of this presnetation will focus on adding hardware loops to GCC in three places; as built-in (intrinsic) funcions, as code-generation patterns for the compiler and within target specific optimization. We shall give special attention to memcpy, the function to copy blocks of memory, and which GCC uses every time a struct is assigned or returned as a result.

To encourage engagement with this project we have provided pre-built binaries, source code, scripts and test results for the comunity.  Additionally, patches are welcome as pull requests against the Open Hardware Group repositories for corev-binutils-gdb and corev-gcc on GitHub.

Mary Bennett is a tool chain engineer at Embecosm, developers of open source compilers, operating systems, models and AI.  She graduated from the University of Surrey with a degree in EEE and a love of Jui Jitsu and climbing.  Mary is best know for her work as Chair of the RISC-V Academia and Training group from 2018 to early 2020. She was winner of the UK Electronic Skills Foundation Scholar of the Year 2019.  Mary is a key engineer in the RISC-V CGEN project and CORE-V Binutils-GDB and GCC projects, and can be found on YouTube speaking on these subjects at numerous conferences over the past three years.

Pietra F. T. Madio moved to the UK from Brazil and has been working as a software engineer for Embecosm since November 2018. Pietra was previously undertaking research and implementation of neural networks using Keras and TensorFlow. She has also written a series of blog posts on implementing face recognition on the Google Coral development board. She currently is a key engineer at the CORE-V Binutils-GDB and GCC project, being the project manager for Binutils-GDB.  Pietra previously presented at the BCS as part of the 2019 Women in Open Source meeting and also spoke at the OSD Forum 2020 about her work on the Core-V Binutils-GDB project.

Jessica Mills is a Software Tool Chain Engineer at Embecosm, developers of open source compilers, operating systems, models and AI. She graduated from the University of Leeds with a masters degree in Electronics and Computer Engineering.  Jessica is a key engineer in both the CORE-V Binutils-GDB and GCC projects, being the project manager for GCC. She has previously presented this work at the OSD Forum 2020.


Lightning Talks and AGM

Join the session on BigBlueButton: https://test.vi4io.org/b/jul-uhk-iht-bvf

This is out annual lightning talk meeting, where members have the opportunity to speak for 10 minutes on their favourite topic.  The meeting will start with the AGM, which all are welcome to attend.

Please register using the link above for a free ticket.

In a first for the BCS, we shall be livestreaming using the open source platform BigBlueButton. We hope this will prove a much richer online experience, particularly for those joining from non-Windows platforms. We shall have a GoToMeeting link available as backup in case of teething problems.  We shall later be posting the videos on our YouTube channel for those who are unable to make it.

AGM

Everyone is welcome to attend, but only BCS members may vote. We will have the brief reports on the past year’s activities and then elect a new committee. Note: non-BCS members may stand for all committee posts except Chair, Treasurer and Membership secretary. So far we have the following nominations, but more are encouraged.

  • Chair: Jeremy Bennett
  • Treasurer: Richard Miller
  • Membership secretary/Advocacy and outreach: Julian Kunkel
  • Inclusion officer: Cornelia Boldyreff
  • Web supremo: Simon Worthington
  • Events sub-committee: Andy Bennett, Sevan Janyan
  • Young Professional/Student representative: Daniel Broomhead
  • Committee members:  Mary Bennett

Existing committee member standing for re-election

We depend on an active committee to put on all our events and drive our advocacy and outreach work. Please consider putting yourself forward.

Adding an instruction to the GNU assembler

Mary Bennett
Pietra F T Madio, @7pietraferreira
Jessica Mills

Over the past couple of years, RISC-V has taken off in both academia and industry. One RISC-V variant, CORE-V, adds extensions which can greatly improve performance and reduce code size. The speakers are leading the project to develop a full GNU tool chain supporting these instruction extensions.

In this lightning talk, they will look at just one part of this project. How to add a new assembly code instruction to the GNU assembler. They’ll use CORE-V hardware loop as the example for this. You’ll learn which files to edit, how to build the new assembler and most importantly how to test it.

A fuller version of this talk will be given at the London RISC-V Meetup (also hosted by BCS Open Source SG) on Monday 19 October, which will look a the complete tool chain. More information can be found in the Open Hardware Group repositories for corev-binutils-gdb and corev-gcc on GitHub.

Mary Bennett is a tool chain engineer at Embecosm, developers of open source compilers, operating systems, models and AI.  She graduated from the University of Surrey with a degree in EEE and a love of Jui Jitsu and climbing.  Mary is best know for her work as Chair of the RISC-V Academia and Training group from 2018 to early 2020. She was winner of the UK Electronic Skills Foundation Scholar of the Year 2019.  Mary is a key engineer in the RISC-V CGEN project and CORE-V Binutils-GDB and GCC projects, and can be found on YouTube speaking on these subjects at numerous conferences over the past three years.

Pietra F. T. Madio moved to the UK from Brazil and has been working as a software engineer for Embecosm since November 2018. Pietra was previously undertaking research and implementation of neural networks using Keras and TensorFlow. She has also written a series of blog posts on implementing face recognition on the Google Coral development board. She currently is a key engineer at the CORE-V Binutils-GDB and GCC project, being the project manager for Binutils-GDB.  Pietra previously presented at the BCS as part of the 2019 Women in Open Source meeting and also spoke at the OSD Forum 2020 about her work on the Core-V Binutils-GDB project.

Jessica Mills is a Software Tool Chain Engineer at Embecosm, developers of open source compilers, operating systems, models and AI. She graduated from the University of Leeds with a masters degree in Electronics and Computer Engineering.  Jessica is a key engineer in both the CORE-V Binutils-GDB and GCC projects, being the project manager for GCC. She has previously presented this work at the OSD Forum 2020.

XO and Sugar on a stick

Professor Cornelia Boldyreff

Cornelia Boldyreff lives in Greenwich and is a part-time Professor at the University of Greenwich in the School of Computing and Mathematical Sciences. She was previously the Associate Dean (Research and Enterprise) at the School of Architecture, Computing and Engineering at the University of East London from 2009 – February 2013.

Cornelia gained her PhD in Software Engineering from the University of Durham where she worked from 1992; she was a Reader in the Computer Science Department when she left.
In 2004 she moved to the University of Lincoln to become the first Professor of Software Engineering at the university, where she co-founded and directed the Centre for Research in Open Source Software.

She has over 25 years’ experience in software engineering research and has published extensively on her research in the field. She is a Fellow of the British Computer Society, and a founding committee member of the BCSWomen Specialist Group, a committee member of the BCS e-Learning Specialist Group, and since 2013 she has chaired the BCS Open Source Specialist Group. She has been actively campaigning for more women in STEM throughout her career.  Together with Miriam Joy Morris and Dr Yasmine Arafa, she founded the start-up, ebartex Ltd, and together they are developing a new digital bartering currency, ebarts.

The HPC Certification Forum: An Open and Free Certification for HPC – A Good Model for OSS?

Dr Julian Kunkel

The diversification of High-Performance Computing (HPC) practitioners challenges the traditional training approaches, which are not able to satisfy the specific needs of users, often coming from non-traditionally HPC disciplines and only interested in learning a particular set of skills. HPC practitioners are expected to have various HPC skills, however, those “skills” have not been well-defined until now. The ability to speak a common language – among HPC educators and users – is critical.

The HPCCF strives to identify this set of competencies for HPC users. It is our aim to thereby facilitate course offers across HPC sites and to provide a certification procedure for HPC practitioners. Ultimately, we aim for the certificates to be recognized and respected by the HPC community and industry. In the competence standard, we already have many open-source skills as open source fuels HPC sytems. We believe this model is scalable and could be applied more in the open-source environment.

Dr Kunkel is a Lecturer in the Department of Computer Science at the University of Reading. He manages several research projects revolving around high-performance computing (HPC) and particularly high-performance storage. He became interested in the topic of HPC storage in 2003, during his studies of computer science.  Julian serves as Membership Secretary and Advocacy and Outreach officer for the BCS Open Source SG.

An Insight into Open Source from a Student’s Perspective.

Daniel Broomhead

The introduction of open source to young people has always been a major downfall for the progression of open source, with young people wanting to get involved with software like Linux but never knowing where to start and how to get involved. Daniel Discusses the experience he had with open source and where he thinks the future need to be heading in order to incorporate the younger generations, reflecting on some of the complications of the past and some reasons why the younger generations struggle to get involved as early as they could have.

Daniel Broomhead is a final year Computer Science Student at the university of Reading. He is the current president of the University’s R. U. Hacking? Society which is a computer Science society with a focus around integrating with a tech and professional environment, including hosting, and running many events like Hackathons.

Daniel is curious about all different areas of computer science and always keen to learn more, however his current passion lies in the cyber security side of things, particularly offensive security, and while he is lacking the current experience and expertise, this is his main focus for the future, and his topic chosen for his final year project.

PCI bus hacking from home – how hard can it be?

Maxim Blinov

Digital protocols are the bread and butter of virtually every electronic engineer today. UART, SPI, I2C, USB, PCIe, the list goes on. But it’s very rarely that we get the opportunity to actually implement one such protocol ourselves.

This project was an excercise to take the role of implementing a protocol, given all the usual limited constraints of a hobbyist operating from home: Create a very simple (but functional) conventional (33MHz) PCI endpoint, visible from Linux, and capable of transferring data from internal FPGA LUTs to Linux Userspace.

The project touched on aspects of FPGA design, design verification, physical debugging approaches, and just a dash (and I really mean just a dash!) of linux kernel modification.

Maxim Blinov is a toolchain engineer working at Embecosm, supporting and customizing GCC toolchains for client-specific architectures and architecture extensions. His main open source work has been benchmarking and customizing the upcoming RISC-V bitmanip extension.

In his free time he dabbles in MCUs, FPGAs, softcore CPUs, and Windows driver development.

A Minimalist Network on Chip on an FPGA

Richard Miller

This is an ongoing project to implement an FPGA NoC (network on chip) which is “as simple as possible but no simpler”. Originally intended as a lightweight alternative to ethernet for linking a cluster of Raspberry Pi Zeroes, it can also serve as the internal communication fabric for a multicore SoC (system on chip).

Richard Miller is a consulting engineer specialising in systems software (operating systems and programming language implementation) and digital logic design. He serves as Treasurer of the BCS Open Source SG.

12 things everyone needs to know about data

Simon Worthington, @51M0NW

In this talk, Simon will give a “Papers We Love”–style walkthrough of a favourite scientific paper from the University of Southampton and the Open Data Institute. Much open data publishing is focused on the act of getting data out of the door safely and correctly, and not many people focus on actually helping others to find and evaluate their datasets. This paper examines the way people search for open data and suggests a 12-point checklist for what potential users need to know that everyone can implement.

Simon is a founder of Register Dynamics, a data consultancy on a mission to make data more useful for everyone, and the creators of Registers.app. As a technologist with years of experience applying cutting-edge data technology to meet real user needs, Simon gets that data is hard. Tell him your data problems on Twitter @RegDyn!

 


Open Source Facial Recognition

This month we have two talks on open source facial recognition.

AGENDA
18:15 – Online session starts

18:30 – Presentations with questions

20:00 – Close

We will be live streaming via GoToWebinar and recording the talks for later posting on YouTube. Details of registration and the livestream will follow shortly.

COVID-19: Contactless Biometrics and Related Technologies for Border Checks On-the-Go

Prof James Ferryman

With COVID-19 hygenic identity verification becomes a crucial issue yet has to be effective and meet travellers¹ and border guards¹ demands on reasonable processing times and privacy. This talk will first present the background in development and deployment of Automated Border Control (ABC), specifically ABC eGates at airports, based on face biometrics. The talk will then briefly present the outcomes of a 4-year EU project called FastPass which sought to develop a harmonised approach to eGates toward more seamless identity confirmation for all border crossing types. The talk will then proceed to propose a vision of the future of ABC whereby eGates are replaced with a no-stopping multimodal biometrics identification system fully taking into account privacy and security issues. In this context, the just-completed EC PROTECT project will be presented including the innovative use cases developed, an overview of the technical solutions, legal and ethical aspects, and outcomes of the demonstrations held.

Prof. Ferryman is a computer scientist and leads the Computational Vision Group within the Department of Computer Science, School of Mathematical, Physical and Computational Sciences (SMPCS), University of Reading. His current research interests include multimodal biometrics, automated video surveillance and benchmarking. He is the author of more than 100 scientific publications. He has participated in a wide range of UK and EU funded research programmes including, under border security, the EU EFFISEC project (FP7-217991) on efficient integrated security checkpoints, the EU IPATCH project (FP7-60756on automated surveillance and decision support system for detection and classification of piracy threats to shipping, and the EU FastPass project (FP7-312583) on development of a harmonised modular reference system for all European automated crossing points, the latter of which he led the largest technological workstream on traveller identification and monitoring. Prof. Ferryman coordinated the EU PROTECT project (H2020-700259, 2016-2019) on exploration of current and future use of biometrics in border security. Prof. Ferryman is a member of the British Computer Society and has acted as the Director of both the British Machine Vision Association and the Security Information Technology Consortium. Since 2000, he has been a Co-Chair of the IEEE International Workshop on Performance Evaluation of Tracking and Surveillance.

AI and Computer Vision

Dr William Jones and Pietra F T Madio

This talk will look at a research project Embecosm has been pursuing on applying facial recognition at the edge using Google’s EdgeTPU and Tensorflow.

William Jones has a research background in computational neuroscience, with a focus on artificial neural networks and machine learning techniques. He leads Embecosm’s AI team, working on applying these machine learning and AI techniques to the domains of compiler and code optimization.

Pietra F. T. Madio is part of Embecosm’s AI team. Recently she has focused on making Artificial Intelligence accessible to teenagers, focusing on Machine Learning and Deep Learning.


OpenChain UK Work Group Inaugural Meeting

BCS Open Source SG has long supported the OpenChain initiative. Most recently we co-hosted a workshop in 2018 with BCS Law SG looking at issues around the Open Source supply chain.

We are pleased to support the inaugural meeting of the OpenChain UK Work Group, which is open to all with an interest in the integrity of Open Source supply chains.  The event is being curated by Orcro Limited supported by Arm.

 

 


An Evening on the theme of FPGAs

Why the J-core open processor is cool

The patents on the bestselling processor of the 1990’s finally expired, and it’s been cloned as open hardware. J-core scales from running SMP Linux down to fitting in a tiny 5000 cell ICE-40 FPGA, and we just uploaded a new open source GPS receiver built around it to github.

Rob Landley maintains the J-core Board Support Package, used to maintain busybox, maintains toybox (the command line utilities used by Android), and used to be the linux kernel documentation maintainer.

Yosys and nextpnr – open source FPGA update

Find out about the newest developments in Yosys and nextpnr, including experimental Xilinx support, synthesis improvements and new place and route algorithms! This talk will also discuss some of the future work
that is planned, with an ambitious target to support 1-million-LUT scale designs on UltraScale+ devices by the end of the year.

David Shah is a self-employed developer working on nextpnr, the open source FPGA place-and-route tool. His previous work also includes Project Trellis, open source bitstream documentation for the Lattice ECP5 FPGAs.

Repurposing Obsolete FPGA-based Products as Development Kits

Commercial FPGA development boards can be relatively cheap or they can have a lot of logic resources and features, but never at the same time.
When commercial products with large FPGAs become obsolete, they can often be found dirt cheap on eBay. With a little bit of effort, some of these products can be reverse engineered and repurposed as generic FPGA development kits. This talk will show some examples, and discuss techniques that can be used to reverse engineer these kind of products.

Tom Verbeure is a hardware engineer at Nvidia where he works on monitor technology. Before Nvidia, he worked on communication ASICs at Barco, Alcatel and Conexant. When he’s not mountain biking in one of the many parks around the SF Bay Area, you can probably find him in his garage, looking through a microscope trying to trace tiny wires on an old PCB, or burning up an LED panel in a misguided attempt to control LED cube with a Cisco WAN modem.


Realising Rewards of Open Data

Open Data is often mentioned alongside its bigger sisters of Open Software and Open Hardware, but so far the movement has not enjoyed as widespead success. But why not? What more could people be getting out of open data that would drive its adoption? In this evening of talks, our speakers make the case for open data and show all the benefits that sharing data openly and widely can bring.

This month we’ll be hosting our evening of talks online. You can join remotely from the comfort of your own home to listen to the speakers and chat in realtime with the other attendees.

AGENDA
18:15 – Join online meeting to chat with other participants
18:30 – Presentations
20:30 – Close

Why Open Data?

Ian Watt

The term Open Data is generally understood to mean data that has been published, often by government, in a neutral format and with a licence that allows re-use. What is less clear are the reasons for opening data: the drivers; the economic and social benefits; the legal obligations and even the moral imperatives. In this short talk Ian Watt will cover these in a way which will not only aid understanding but will help someone to build a case to make their organisation’s data open too.

Ian is a Data Scientist and CEO and co-founder of the charity Code The City which operates Scotland’s only node of the Open Data Institute – ODI Aberdeen. He has been a prominent advocate for Open Data for almost ten years. Until his retirement from local government in 2017 he led the open data programme for Scotland’s seven cities. He recently became the civic society lead for Commitment Three (covering open data and data literacy) for the Open Government Scotland plan. He is on the steering group of Stirling University’s Data Commons research project, and is a non-executive director of Democracy Club, a UK CIC who use open data to improve democratic engagement.

Comparing Open Data With Open Source

Leigh Dodds

What is open data? How has the publishing and use of open data evolved in different communities? And how does it compare to the open source movement?

In this talk Leigh will explore how the open data movement has evolved, how it compares with other “open” approaches and discuss how combining approaches can help to create open, trustworthy data ecosystems.

Leigh is Director of Delivery at the Open Data Institute where he helps to define, lead and deliver a range of projects and programmes to help build an open, trustworthy data ecosystem. Leigh has over 20 years experience in a variety of technology focused roles, including as a software engineer, product manager and technical consultant. For the last ten years he has worked on a range of open data projects across the public and private sector. He also chairs the Bath: Hacked volunteer group, exploring use of open data in Bath & North East Somerset.

Open property data for decision-making and innovation

Natalie Record

Open data relating to the built environment is hard to find, use and trust. When found, data is rarely standardised which makes it difficult to link to other data sets and build up a holistic picture, both nationally and at a hyperlocal level. In this talk, Natalie will explore why there is a pressing need to increase access to unique open identifiers and geospatial data sets (non-personal) relating to places and property to inform decision-making and enable the emerging PropTech sector to innovate new solutions to help overcome housing and planning challenges.

Natalie leads on Digital Housing and Planning Policy at the Ministry of Housing, Communities and Local Government in the Digital Land team. The team works in the open to help local government, policymakers and the PropTech sector to get better access to trusted land, planning and housing data to build new services. Natalie has a background in Urban Planning and co-founded a PropTech startup, Walulel, in 2016 before moving on to work on the wider digital transformation of planning through policy roles in local and central government.


Open Source in Training and Education

Open source code is widely used in education and training events. On this thematic evening, we’ll look into various aspects.

AGENDA
18:15 – Join online meeting to chat with other participants

18:30 – Short introduction (5 min) of the evening by Julian Kunkel and Marry Bennett

18:35 – Presentations

20:35 – Closing Discussion

We were live streaming via BigBlueButton and recording the talks for later posting on YouTube.

 

The videos are available in the playlist here.

Digital freedom with youth and education

Video Presentation

The goals of the schul-frei project are to offer schools and other educational establishments free software. It shows the alternatives to proprietary educational solutions from big tech companies. We collect different single solutions which can take on their specific task and put them together, so that we can offer schools one single solution. It contains a learning management system (Moodle), an app for timetables (AlekSIS) or an operating system, specially made for educational usage (DebianEdu) and more.
The opportunities of free software are the independence of companies, the freedom to decide how the software is used in schools and transparent data protection. We also want to include the youth into developing software for their schools. Free software in schools allows students and teachers to develop together apps for their school, which can be integrated into the running system we present.
Benedict Suska, Kirill Schmidt and Domink George are members of Teckids, an association of which goals are to bring free software to youth and education.

Open Source in Teaching RISC-V

Video Presentation

RISC-V is an open instruction set architecture, attracting both open source and proprietary processor implementations. This makes it very attractive for education in computer organization and other courses. In this talk I will present our efforts to use and share open source implementations and tools for educational purposes.

Stefan Wallentowitz is Professor at Munich University of Applied Sciences, on the board of FOSSi Foundation and RISC-V.

 

Championing Open Source in Higher Education

Video Presentation

At the University of Edinburgh our Content Management System is based on Drupal and we try to support this by running code sprints, sponsoring events and giving venues space. However, we do find it hard to contribute code and are exploring ways to improve this. We also work with OpenUK and want to support running an event for UK University academics. This is a course aimed to help academics understand ‘open source participation and the curricular and pedagogical considerations of teaching open source’.

Bruce Darby is Product Owner for the University of Edinburgh’s Content Management System which is based on the Open Source CMS Drupal.

Open HPC Certification and Relationship to OSS

Video Presentation

The diversification of High-Performance Computing (HPC) practitioners challenges the traditional training approaches, which are not able to satisfy the specific needs of users, often coming from non-traditionally HPC disciplines and only interested in learning a particular set of skills. HPC practitioners are expected to have various HPC skills, however, those “skills” have not been well-defined until now. The ability to speak a common language – among HPC educators and users – is critical.

The HPCCF strives to identify this set of competencies for HPC users. It is our aim to thereby facilitate course offers across HPC sites and to provide a certification procedure for HPC practitioners. Ultimately, we aim for the certificates to be recognized and respected by the HPC community and industry. In the competence standard, we already have many open-source skills as open source fuels HPC systems. In this talk, technical details of how the certification uses open source software is presented.

Julian Kunkel is a Lecturer in the Department of Computer Science at the University of Reading. He manages several research projects revolving around high-performance computing (HPC) and particularly high-performance storage. He became interested in the topic of HPC storage in 2003, during his studies of computer science.  Julian serves as Membership Secretary and Advocacy and Outreach officer for the BCS Open Source SG. He is currently the chair for the HPC Certification Forum (HPC-CF).

 

 

Note: Please aim to connect at the latest by 18:25 as the event will start at 18:30 prompt.


Open source software for scientific and parallel computing

https://www.youtube.com/playlist?list=PLxde5XJWZRbR5pmLiMRp771cvvfQoxAhvIn public data centers and in computational science, open-source software plays a key role to create a productive environment for researchers.

Computational science is the modeling and simulation of the laws of nature within computer systems that offer a well-defined environment for experimental investigation. Models for climate, protein folding, or nanomaterials, for example, can be simulated and manipulated at will without being restricted by the laws of nature, and scientists no longer have to conduct time-consuming and error-prone experiments in the real world. This method leads to new observations and understandings of phenomena that would otherwise be too fast or too slow to comprehend in vitro. The processing of observational data like sensor networks, satellites, and other data-driven workflows is yet another challenge as it usually dominated by the input/output of data.

Complex climate and weather simulations can have 100.000 to million lines of codes and must be maintained and developed further for a decade at least. Therefore, scientific software is mostly open-source, particularly for large scale simulations and bleeding-edge research in a scientific domain.

This month we’ll be hosting our evening of talks online. You can join remotely from the comfort of your own home to listen to the speakers and chat in realtime with the other attendees.

AGENDA
18:15 – Join online meeting to chat with other participants

18:30 – Short introduction (5 min) of the evening by Julian Kunkel — Slides

18:35 – Presentations

20:35 – Close

We were live streaming via GoToWebinar and recording the talks for later posting on YouTube.

The videos are now published on YouTube and slides are linked below.

 

Open Source Software in High-Performance Computing

Shane Canon, Lawrence Berkeley National Laboratory/NERSC, USA

High-Performance Scientific Computing is heavily dependent on a rich ecosystem of open-source software.  The HPC community is both a consumer and a contributor to the broader open-source community.  In this presentation, we will review the evolution of open-source software use in HPC, give examples of how the HPC community has contributed to its growth, and the future direction of open-source in HPC.

Some lessons learned from creating and using the Ceph open-source storage system

Carlos Maltzahn, UC Santa Cruz, USA
In 2005 at UC Santa Cruz, Sage Weil created a fully functional parallel file system we called Ceph. In 2006 we open-sourced Ceph under the LGPLv2 license. After his graduation in 2007, Sage was able to continue working on Ceph, built a community around it, got the Ceph client accepted into the Linux kernel in 2010, achieved production readiness in 2012, founded the startup Inktank in 2012, and sold it to Red Hat in 2014 for $175m. Sage then donated $2m to UC Santa Cruz to create a structure that would help other students to have a similar career at the university as he had. In 2015 I kicked off the Center for Research in Open Source Software (CROSS). Meanwhile, starting in 2009 I built a research program around programmable storage systems and decided to use Ceph as the primary research prototyping platform. As Ceph became more widely used, that choice gained a positive impact on the job prospects of our students, on the generality of our research results, on the funding prospects of our research, and on the pace of delivering research.
In this talk, I will report on some of the lessons learned, including how our use of Ceph as a research platform evolved and why an open-source storage system like Ceph can fundamentally change the dynamics of research and entire industries. I will end with a quick overview of the SkyhookDM project (skyhookdm.com).
Bio: Dr. Carlos Maltzahn is the founder and director of the UC Santa Cruz Center for Research in Open Source Software (CROSS). Dr. Maltzahn also co-founded the Systems Research Lab, known for its cutting-edge work on programmable storage systems, big data storage & processing, scalable data management, distributed system performance management, and practical reproducible evaluation of computer systems. Carlos joined UC Santa Cruz in 2004, after five years at Netapp working on network-intermediaries and storage systems. In 2005 he co-founded and became a key mentor on Sage Weil’s Ceph project. In 2008 Carlos became a member of the computer science faculty at UC Santa Cruz and has graduated nine Ph.D. students since. Carlos graduated with a M.S. and Ph.D. in Computer Science from University of Colorado at Boulder.

High Performance Computing in a world of Data Science

Martin Callaghan, University of Leeds, UK

Slides as PDF

Universities and other research organisations have developed and used High Performance Computing (HPC) systems for a number of years to support problems solving across many computational domains including Computational Fluid Dynamics and Molecular Dynamics.  Their design features such as a batch processing system and a fast interconnect make them ideal to support these often highly parallel tools and applications.

In recent years though, with the increased interest in Data Science across a number of research fields, HPC has found itself in the position of having to support quite different tools and methodologies.

In this talk, I’ll discuss the design journey we have taken for our institutional HPC, some of the Open Source projects, tools and techniques we use with our research colleagues to support Data Science problems and some of our plans for the future.

Martin Callaghan Research Computing Manager and lead the Research Software Engineering team at the University of Leeds in the UK, where we provide High Performance Computing (HPC), Programming and Software Development consultancy across a diverse research community. My role involves Research Software Engineering, training, consultancy and outreach.

He also manages a comprehensive HPC and Research Computing training programme designed to be a ‘zero to hero’ structured introduction to HPC, Cloud and research software development.

Before joining the University of Leeds, he worked as an Engineer designing machine tool control systems, a teacher and ran my own training and consultancy business. Personal research interests are in text analytics, particularly using neural networks to summarise text at scale.

 

Note: Please aim to connect at the latest by 18:25 as the event will start at 18:30 prompt.