Reflections on ORConf 2015

Us ICCLab folk are always interested in new ideas, particularly those that could have a profound impact on computing in general and cloud computing in particular. Consequently, we couldn’t miss out on the opportunity of attending ORConf – a conference loosely centred around open source silicon – which was free and (more or less) just down the road at CERN.

The conference itself was superb, comprising of an excellent mix of hobbyists/open source advocates, industry folks and academics with some of the people wearing more than one hat. There was also quite a diverse set of backgrounds ranging from ASIC designers to FPGA guys to compiler designers to some simpler software types. The quality of people was overwhelming with excellent guys from high profile organizations such as Intel, Google, Qualcomm, nvidia, Uni Cambridge, EPFL, ETH and Berkeley (although many of the industry folk were not specifically representing their employers).

The conference has been growing steadily over the last number of years, reaching about 100 people this year. The single track event covered a very diverse range of topics. Unfortunately, we missed the OpenRISC session on Friday which had a very strong RISC-V dominance and was very interesting by all reports (must catch up on this content later!). The content on Saturday and Sunday addressed many issues most of which involved demonstrating tools of some description: Clifford Wolf’s trojan work on Yosys, his open-source Verilog synthesis tool was remarkable and we also liked Chris Higg’s work on raising abstraction levels of verification frameworks and realizing them in python through the CocoTB framework. Michael Gielda of Antmicro’s presentation on the Axiom Gamma open camera project was very interesting and looked like a really fun EU project with real value-creation potential and the Cosmic Pi presentation by James Dilwyn Devine was another fun project on gathering cosmic ray data using open hardware designs. There were other interesting contributions including some on open source simulation tools, debugging solutions and packaging hardware and software together.

Somewhat ironically, it seemed as though the presentation by Guarav Kaul of Intel was the most talked about contribution. Guarav provided lots of interesting detail on how Intel’s new combined Xeon/FPGA chips are designed, how they can be programmed and integrated into applications. Interestingly, the FPGAs can access data over the Intel QPI interface and consequently, they can directly access the main system memory which is obviously much faster and more energy efficient than needing to go across the PCIe bus. The programming model, however still seems quite basic with little in the way of Operating System supports or virtualization supports for the new FPGA functionalities: obviously, this is something that must be being worked on.

Two more interesting points are of note: a foundation – the Free and Open Source Silicon Foundation – is being formed to try to harness the energy of this community and propagate the message around open source silicon and ultimately grow the community. There was some discussion around how to do this although we did feel that the concrete near term goals of the foundation were not sufficiently clear at the end of the discussion. It was generally agreed that two specific areas that the foundation could serve a useful purpose were around maintaining a catalog of relevant projects and offering recommendations for licensing of open source silicon. The second (related) noteworthy point was specifically around appropriate licensing models for open source silicon and whether the well known open source software licenses are appropriate for this context. Unfortunately, we had to leave before this discussion got going, but the presence of significant legal expertise in the room surely facilitated a useful and interesting discussion.

Overall, this was one of the best conferences we’ve attended in some time: the innovation in this field is sure to impact computing generally and most certainly in the Data Centre. Hat tip to Julius Baxter and team for organizing a super event and we’re looking forward to next year already!

Leave a Reply

Your email address will not be published. Required fields are marked *