Brad Penoff

Brad Penoff

Brad comes from Ohio where he attended The Ohio State University as an undergraduate in Computer Science & Engineering as well as Computational Linguistics. After 3 years at Sun Microsystems, he attended the University of British Columbia where he obtained both a Masters and PhD in Computer Science where he focused on the implementation of large-scale parallel compute clusters, spending two summer internship at Argonne National Laboratory's supercomputing group. He has worked at Google working on the Scale & Performance team for Payments since Fall 2011, and leads the Throughput team.

Research Areas

Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    Portable and Performant Userspace SCTP Stack
    Alan Wagner
    Michael Tuexen
    Irene Ruengeler
    Computer Communications and Networks (ICCCN), 2012 21st International Conference on, IEEE
    Preview abstract One of only two new transport protocols introduced in the last 30 years is the Stream Control Transmission Protocol (SCTP). SCTP enables capabilities like additional throughput and fault tolerance for multihomed hosts. An SCTP implementation is included with the Linux kernel and another implementation called sctplib functions successfully in userspace on several platforms but unfortunately neither of these implementations have all of the latest features nor do they perform as well as the FreeBSD kernel implementation of SCTP. We were motivated to produce a portable implementation of the FreeBSD kernel SCTP stack that operates in userspace of any system because of both our desires to obtain a higher performance SCTP stack for Linux as well as to exploit recent developments in hardware virtualization and transport protocol onloading. Unlike any other userspace transport implementation for TCP or SCTP, our userspace SCTP stack simultaneously achieves similar throughput and latency as the Linux kernel TCP stack, without compromising on any of the transport's features as well as maintaining true portability across multiple operating systems and devices. We create a callback API and implement a threshold to control its usage; our userspace SCTP stack with these optimizations obtains higher throughput than the Linux kernel implementation of SCTP. We describe our userspace SCTP stack's design and demonstrate how it gives similar throughput and latency on Linux as the kernel TCP implementation, with the benefits of the new features of SCTP. View details
    A new fast algorithm for connecting the INET simulation framework to applications in real-time
    Irene Rüngeler
    Michael Tüxen
    Alan Wagner
    SimuTools(2011), pp. 75-80
    Employing transport layer multi-railing in cluster networks
    Humaira Kamal
    Alan Wagner
    Mike Tsai
    Karol Mroz
    Janardhan R. Iyengar
    J. Parallel Distrib. Comput., 70(2010), pp. 259-269
    MPI-NeTSim: A Network Simulation Module for MPI
    Alan Wagner
    Michael Tüxen
    Irene Rüngeler
    ICPADS(2009), pp. 464-471
    A Hybrid MPI design using SCTP and iWARP
    Mike Tsai
    Alan Wagner
    IPDPS(2008), pp. 1-8
    Using CMT in SCTP-Based MPI to Exploit Multiple Interfaces in Cluster Nodes
    Mike Tsai
    Janardhan R. Iyengar
    Alan Wagner
    PVM/MPI(2007), pp. 204-212
    Towards MPI progression layer elimination with TCP and SCTP
    Alan Wagner
    IPDPS(2006)
    Using SCTP to hide latency in MPI programs
    Humaira Kamal
    Mike Tsai
    E. Vong
    Alan Wagner
    IPDPS(2006)
    SCTP versus TCP for MPI
    Humaira Kamal
    Alan Wagner
    SC(2005), pp. 30
    Evaluating transport level protocols for MPI in the Internet
    Humaira Kamal
    Alan Wagner
    Communications in Computing(2005), pp. 71-77
    SCTP-Based Middleware for MPI in Wide-Area Networks
    Humaira Kamal
    Alan Wagner
    CNSR(2005), pp. 157-162