With the slowdown of Moore’s law as we know it, the Cloud is resorting to heterogeneous, accelerated computing to satisfy the ever-increasing demand for performance and power efficiency. In just a few years, FPGAs have emerged as compute accelerators next to GPUs and are part of the standard offerings from many Cloud vendors. However, the development environment, deployment procedures, security measures, and monitoring tools are different for each platform and the portability of the FPGA kernel designs remains limited.
In this workshop, leading platform providers, designers of development environments, and developers are going to present the state-of-the-art for Cloud FPGA platforms and explore opportunities and directions for future improvements from the developer’s point of view. Instead of focusing on the performance and optimization of a specific application, the goal of this workshop is to highlight the challenges, which a Cloud application developer faces when designing, implementing, deploying, scaling and debugging Cloud services on Cloud FPGA platforms. A focus area for this years edition are end-to-end tool chains, compilation and debugging tools for distributed FPGA platforms.
We thank all speakers for their excellent presentations and all participants for joining and their interesting questions. We felt we had a very insightful workshop with a lively debate and we also were happy to see the convergence of different approaches during our discussions. The slides and recordings of most talks are linked below.
We hope to see all of you next year at FPL again personally, where we might organize another cFDevOps workshop. Stay safe!
The workshop will take place on Monday, August 30, 14:30h - 18:30h CEST. The workshop will be held online (Zoom Webinar/Whova platform).
Once registered, you can find the access details here.
|Session 1||Development and Debugging of Cloud FPGA applications|
|14:40||On the development of distributed applications with Intel FPGA SDK for OpenCL||Tiziano De Matteis, ETH Zürich|
|15:10||And, what about debugging for multi-FPGA systems?||Paul Chow, University of Toronto|
|15:40||climbing EVEREST: dEsign enVironmEnt foR Extreme-Scale big data analyTics on heterogeneous platforms||Christian Pilato, Politecnico di Milano, Michele Paolino, Virtual Open Systems|
|16:10||Q & A||Everyone|
|Session 2||Operation of Cloud FPGA deployments|
|16:40||Single-line deployment of Cloud FPGAs||Chris Kachris, InAccel|
|17:10||AQUA (Advanced Query Accelerator) for Amazon Redshift||Andrew Caldwell, AWS Redshift|
|17:40||XACC-ETHZ cluster||He Zhenhao, ETH Zürich|
|18:10||Q & A||Everyone|
The registration is closed.
More details about FPL 2021 can be found here.
by Tiziano De Matteis, ETH Zürich, Scalable Parallel Computing Laboratory
Abstract: In this talk, we will share our experiences in developing a distributed communication library for multi-FPGA systems (Streaming Messages Interface) and distributed applications that build on top of it. We will present the main challenges that we had to face in developing and emulating such applications, together with our adopted solutions. Finally, we will briefly discuss what we believe are necessary tools to further enable the development/debugging of distributed FPGA programming.
by Paul Chow, Professor Computer Engineering & Electronics, University of Toronto
Abstract: In reality, very few beyond Microsoft have built significant multi-FPGA systems. This means that very few have even thought about how to debug a single application running on multiple FPGAs because the need has not arisen. Most FPGA designers are familiar with tools such as Intel’s SignalTap and the Xilinx ILAs, but they have been mostly debugging one FPGA at a time. Would you really want to use these tools on an application running on 16 network-connected FPGAs in the cloud? Can such tools even work in that environment? In this talk I will present work done at the University of Toronto where we have considered this problem and developed some early-stage techniques to help find bugs and do performance analysis of a multi-FPGA application.
by Christian Pilato, Assistant Professor, Dipartimento di Elettronica, Politecnico di Milano and Michele Paolino, Virtual Open System
Abstract: Modern big data applications demand heterogeneous platforms to achieve high performance and energy efficiency. FPGA acceleration brings significant advantages but requires complex hardware skills to obtain efficient designs, especially regarding data movements. IBM cloudFPGA is a paradigmatic example: It offers an infrastructure to scale FPGA acceleration to multiple nodes, but the optimization of the user’s code is left to the designer. This talk will introduce EVEREST, a project that complements cloudFPGA solutions to tackle such challenges. The EVEREST approach aims at providing a design environment that operates at both design and run times. At design time, it offers a data-driven programming environment that combines domain-specific extensions, compiler optimizations, and high-level synthesis to create efficient FPGA accelerators along with complex but flexible memory architectures. It also features a runtime environment to virtualize and manage the available hardware resources and dynamically tune the application execution accordingly.
by Chris Kachris, CEO of InAccel
Abstract: How can users deploy and scale their FPGA-based application instantly on the cloud and in data centers? How can we eliminate the complexity of FPGA and allow seamless deployment as easy as the CPUs? In this talk we will show how users can deploy and scale their FPGA applications on a Kubernetes cluster as easy as if it was typical processors. We will show how users can seamlessly invoke the accelerated functions on a cluster of FPGAs and how multiple users can share the available FPGA resources easier than ever. Finally we will show how users can perform DevOps on FPGAs on the cloud with an integrated framework.
by He Zhenhao, ETH Zürich, Institute for Computing Platforms - Systems Group
Abstract: In this talk I will present a short overview of the XACC-ETHZ cluster, its operation, and briefly describe some of the work being done on it. The cluster has been in operation since the beginning of 2020 and currently hosts more than 100 researchers from 25 different institutions and almost as many countries. In the talk I will present the hardware, the available software and set up, as well as the mode of operation.
by Andrew Caldwell, AWS Redshift
Chris Kachris the founder and CEO of InAccel that helps companies speedup their applications using hardware accelerators (FPGAs) in the cloud or on-prem. Inaccel, using a unique FPGA orchestrator, allows scalable deployment of FPGA clusters. Chris holds a Ph.D. from Delft University of Technology. He has published more than 70 papers in international journals and conferences with more than 1800 citations. He has over 15 years of experience on FPGAs and he is the editor of the book Hardware Accelerators in Data Centers.
Christoph Hagleitner leads the cloudFPGA project at the IBM Research Europe Lab (ZRL) in Ruschlikon, Switzerland. He obtained a Ph.D. degree for a thesis on CMOS-integrated Microsensors from ETH, Zurich, Switzerland in 2002. In 2003 he joined IBM Research to work on the system architecture of a novel probe-storage device (“millipede”-project). In 2008, he started to build up a new research group in the area of accelerator technologies. The team initially focused on integrated accelerator cores and gradually expanded its research to heterogeneous computing systems and their applications.
Dionysios Diamantopoulos is a Research Scientist in the Cloud & AI Systems Research department of IBM Research Europe. Dionysios holds a D.Eng. from CEID/University of Patras (2009) and a PhD from ECE/Technical University of Athens (2015). His research is disseminated in one book chapter and more than 30 publications in international conferences and journals. He is the co-inventor of four filed patents and one pending (USPTO). His research interests lie in the system-level specialization of heterogeneous platforms for the hybrid cloud era. He is a member of HiPEAC, IEEE, OpenPOWER and Technical Chamber of Greece.
Burkhard Ringlein is a Predoctoral Researcher in the Cloud & AI Systems Research department of the IBM Research Zurich Laboratory and pursues his PhD in cooperation with the Department of Computer Science, Computer Architecture of the Friedrich-Alexander University Erlangen-Nürnberg, Germany. As part of the cloudFPGA project, he is focusing on Distributed Reconfigurable Architectures in the context of High-Performance-Computing and AI Acceleration.
Christian Plessl is Professor for Computer Science of Paderborn University and director of the Paderborn Center for Parallel Computing (PC²). His research is focussed on reconfigurable architectures, compilation and runtime systems and applications. Leveraging the longstanding expertise in FPGA acceleration at Paderborn University, PC² has for the first time deployed FPGAs in an HPC production system (Noctua) in 2018. This installation serves as a development platform for porting HPC codes and libraries to FPGAs.