skip to main content

Title: Introducing Beginners to Distributed Computing using Raspberry Pi Clusters
The 2019 ABET computer science criteria requires that all computing students learn parallel and distributed computing (PDC) as undergraduates, and CS2013 recommends at least fifteen hours of PDC in the undergraduate curriculum. Consequently, many educators look for easy ways to integrate PDC into courses at their institutions. This hands-on workshop introduces Message Passing Interface (MPI) basics in C/C++ and Python using clusters of Raspberry Pis. The Message Passing Interface (MPI) is a multi-language, platform independent, industry-standard library for parallel and distributed computing. Raspberry Pis are an inexpensive and engaging hardware platform for studying PDC as early as the first course. Participants will experience how to teach distributed computing essentials with MPI by means of reusable, effective "parallel patterns", including single program multiple data (SPMD) execution, send-receive message passing, the master-worker pattern, parallel loop patterns, and other common patterns, plus longer "exemplar" programs that use MPI to solve significant applied problems. The workshop includes: (i) personal experience with the Raspberry Pi (clusters provided for workshop use); (ii) assembly of Beowulf clusters of Raspberry Pis quickly in the classroom; (iii) self-paced hands-on experimentation with the working MPI programs; and (iv) a discussion of how these may be used to achieve the goals more » of CS2013 and ABET. No prior experience with MPI, PDC, or the Raspberry Pi is expected. All materials from this workshop will be freely available from; participants should bring a laptop to access these materials. « less
; ; ;
Award ID(s):
Publication Date:
Journal Name:
SIGCSE '20: Proceedings of the 51st ACM Technical Symposium on Computer Science Education
Page Range or eLocation-ID:
1388 to 1388
Sponsoring Org:
National Science Foundation
More Like this
  1. The ACM/IEEE CS 2013 report recommends fifteen hours of parallel & distributed computing (PDC) education for every undergraduate. This workshop illustrates the use of the Raspberry Pi as an inexpensive, multicore platform for teaching shared-memory parallel programming. The inexpensive and tactile nature of the Raspberry Pi enables each student to experience her own parallel multiprocessor through sight and touch. In this hands-on workshop, we will teach attendees how they can leverage the Raspberry Pi and the OpenMP library to teach shared-memory parallel concepts in their own classrooms. All CS educators who are interested in learning about the Raspberry Pi, sharedmore »memory parallelism, and OpenMP are encouraged to attend. In Part I of the workshop, each participant will connect to and learn about the Raspberry Pi's multicore capabilities. In Part II, each participant will engage in self-paced, hands-on exploration of basic parallel computing concepts using the OpenMP "patternlets" from In Part III, participants will investigate more complex applications, such as numeric integration and drug design and study how these applications can be parallelized using OpenMP. We will conclude the workshop with a series of lightning talks discussing how the Raspberry Pi has been used to teach parallel computing concepts at different institutions. We will also present a summary of student perceptions of the Raspberry Pi. All materials from this workshop will be freely available from Space is limited to 20 participants. A laptop is required.« less
  2. Teaching parallel and distributed computing (PDC) concepts is an ongoing and pressing concern for many undergraduate educators. The ACM/IEEE CS Joint Task Force on Computing Curricula (CS2013) recommends 15 hours of PDC education in the undergraduate curriculum. Most recently, the 2019 ABET Criteria for Accrediting Computer Science requires coverage of PDC topics. For faculty who are unfamiliar with PDC, the prospect of incorporating parallel computing into their courses can seem very daunting. For example, should PDC concepts be covered in a single required course (perhaps computer systems) or be scattered throughout different courses in the undergraduate curriculum? What languages aremore »the best/easiest for students to learn PDC? How much revision is truly needed? This Birds of a Feather session provides a platform for computing educators to discuss the common challenges they face when attempting to incorporate PDC into their curricula and share potential solutions. Chiefly, the organizers are interested in identifying "gap areas" that hinder a faculty member's ability to integrate PDC into their undergraduate courses. The multiple viewpoints and expertise provided by the BOF leaders should lead to lively discourse and enable experienced faculty to share their strategies with those beginning to add PDC across their curricula. We anticipate that this session will be of interest to all CS faculty looking to integrate PDC into their courses and curricula.« less
  3. The NSF/IEEE-TCPP Parallel and Distributed Computing curriculum guidelines released in 2012 (PDC12) is an effort to bring more parallel computing education to early computer science courses. It has been moderately successful, with the inclusion of some PDC topics in the ACM/IEEE Computer Science curriculum guidelines in 2013 (CS13) and some coverage of topics in early CS courses in some universities in the U.S. and around the world. A reason often cited for the lack of a broader adoption is the difficulty for instructors who are not already knowledgable in PDC topics to learn how to teach those topics and alignmore »their learning objectives with early CS courses. There have been attempts at bringing textbook chapters, lecture slides, assignments, and demos to the hands of the instructors of early CS classes. However, the effort required to plow through all the available materials and figure out what is relevant to a particular class is daunting. This paper argues that classifying pedagogical materials against the CS13 guidelines and the PDC12 guidelines can provide the means necessary to reduce the burden of adoption for instructors. In this paper, we present CAR-CS, a system that can be used to categorize pedagogical materials according to well- known and established curricular guidelines and show that CAR-CS can be leveraged 1) by PDC experts to identify topics for which pedagogical material does not exist and that should be developed, 2) by instructors of early CS courses to find materials that are similar to the one that they use but that also cover PDC topics, 3) by instructors to check the topics that a course currently covers and those it does not cover.« less
  4. It is typical in High Performance Computing (HPC) courses to give students access to HPC platforms so that they can benefit from hands-on learning opportunities. Using such platforms, however, comes with logistical and pedagogical challenges. For instance, a logistical challenge is that access to representative platforms must be granted to students, which can be difficult for some institutions or course modalities; and a pedagogical challenge is that hands-on learning opportunities are constrained by the configurations of these platforms. A way to address these challenges is to instead simulate program executions on arbitrary HPC platform configurations. In this work we focusmore »on simulation in the specific context of distributed-memory computing and MPI programming education. While using simulation in this context has been explored in previous works, our approach offers two crucial advantages. First, students write standard MPI programs and can both debug and analyze the performance of their programs in simulation mode. Second, large-scale executions can be simulated in short amounts of time on a single standard laptop computer. This is possible thanks to SMPI, an MPI simulator provided as part of SimGrid. After detailing the challenges involved when using HPC platforms for HPC education and providing background information about SMPI, we present SMPI Courseware. SMPI Courseware is a set of in-simulation assignments that can be incorporated into HPC courses to provide students with hands-on experience for distributed-memory computing and MPI programming learning objectives. We describe some these assignments, highlighting how simulation with SMPI enhances the student learning experience.« less
  5. The Message Passing Interface (MPI) has been the dominant message passing solution for scientific computing for decades. MPI point-to-point communications are highly efficient mechanisms for process-to- process communication. However, MPI performance is slowed by concurrency protections in the MPI library when processes utilize multiple threads. MPI’s current thread-level interface imposes these overheads throughout the library when thread safety is needed. While much work has been done to reduce multithreading overheads in MPI, a solution is needed that reduces the number of messages exchanged in a threaded environment. Partitioned communication is included in the MPI 4.0 standard as an alternative thatmore »addresses the challenges of multithreaded communication in MPI today. Partitioned communication reduces overall message volume by creating a buffer-sharing mechanism between threads such that they can indicate when portions of a communication buffer are available to be sent. Separation of the control and data planes in MPI is enabled by allowing persistent initialization and single occurrence message buffer matching from the indication that the data is ready to be sent. This enables the usage commands (destination, size, etc.) can be set up prior to data buffer readiness with readiness triggered by a simple doorbell/counter later. This approach is useful for future development of MPI operations in environments where traditional networking commands can have performance challenges, like accelerators (GPUs, FPGAs). In this paper,we detail the design and implementation of a layered library (built on top of MPI-3.1) and an integrated Open MPI solution that supports the new, MPI-4.0 partitioned communication feature set. The library will enable applications to use currently released MPI implementations and older legacy libraries to provide partitioned communication support while also enabling further exploration of this new communication model in new applications and use cases. We will compare the designs of the library and native Open MPI support, provide performance results and comparisons between the two approaches, and lessons learned from the implementation of partitioned communication in both library and native forms. We find that the native implementation and library have similar performance with a percentage difference under 0.94% in microbenchmarks and performance within 5% for a partitioned communication enabled proxy application.« less