ACM SIGCOMM 2018 Afternoon Workshop on Kernel Bypassing Networks (KBNets 2018)
Monday, August 20, 2018, InterContinental
2:00 pm - 2:05 pm Opening
2:05 pm - 2:40 pm Keynote I: RDMA-Based Networking Technologies and Middleware for Next-Generation Clusters and Data Centers
Speaker: D. K. Panda (OSU, USA)
2:40 pm - 3:15 pm Keynote II: To RDMA and Beyond
Speaker: Eitan Zahavi (Mellanox, Israel)
3:15 pm - 3:45 pm Tea/Coffee Break
Location: InterContinental Pre-Function Area
- Tea/Coffee Break
3:45 pm - 5:25 pm Technical Session
- Technical Session
3:45 pm - 4:10 pm
Circuit Switched VM Networks for Zero-Copy IO
Johannes Krude, Mirko Stoffers, Klaus Wehrle (RWTH, Germany)
4:10 pm - 4:35 pm
A Formally Verified NAT Stack
Solal Pirelli, Arseniy Zaostrovnykh, George Candea (EPFL, Switzerland)
4:35 pm - 5:00 pm
Flow control for Latency-Critical RPCs
Marios Kogias, Edouard Bugnion (EPFL, Switzerland)
5:00 pm - 5:25 pm
Hash Table Design and Optimization for Software Virtual Switches
Yipeng Wang, Sameh Gobriel, Ren Wang, Tsung-Yuan Charlie Tai, Cristian Dumitrescu (Intel, USA)
5:25 pm - 5:30 pm Closing
Call for Papers
Kernel Bypassing Networks (including, but not limited to RDMA, DPDK and SmartNIC) have recently drawn much attention from both the research community and the industry. Emerging applications such as AI training, distributed storage systems, virtual networking, and software middleboxes/NFV demand low latency, high bandwidth and low CPU overhead from the network, and have been shown to benefit significantly from technologies that bypass the conventional OS network stack. At the same time, recent switch and NIC developments (e.g., RoCE) have paved the way to the largescale deployment of KBNets.
KBNets are thus starting to be deployed in practice. But plenty of research remains to be done. For example, the research community is still debating whether the kernel should be bypassed in different usecases, what sort of control plane and network management systems are needed for such networks, and how to design congestion control protocols for KBNets and deal with inherent problems such as deadlocks, and how to design programming languages for KBNets. Perhaps more importantly, there is much more work needed to understand how KBNets will impact the design of distributed systems and applications that run over these networks.
We believe that networking community must expedite the research on the design of kernel bypassing networks, understand both their strengths and their weaknesses, and rethink how we design distributed systems to take advantage of these networks.
The ACM SIGCOMM Workshop on Kernel Bypassing Networks (KBNets 2018) is organized with the goal of bringing together researchers from the networking, operating systems, and distributed systems communities to promote the development and evolution of kernel bypassing networks. All submissions related to KBNets and KBNetsbased systems, including network/system architecture, design, implementation, programing language, simulation, modeling, analysis, and measurement will be welcome. We highly encourage novel and innovative early stage work that will encourage discussion and future research on KBNets.
Topics of Interest
Topics include but are not limited to:
- Network transport for kernel bypassing networks
- Control plane for kernel bypassing networks
- Security issues regarding kernel bypassing networks
- Distributed systems that are based on kernel bypassing networks, e.g., AI training, distributed storage, database and inmemory caches
- Data center network architectures for kernel bypassing networks
- Virtualization for kernel bypassing networks
- Programming languages for kernel bypassing networks
- NIC/switch hardware design for kernel bypassing networks
- Middleboxes/NFV optimization with kernel bypassing networks
- Diagnosing and troubleshooting kernel bypassing networks
- Experiences and bestpractices in deploying kernel bypassing networks
- Measurements of kernel bypassing and kernel optimizing networks
- Performance studies of kernel bypassing networks and applications
- Transition and backward compatibility with traditional network stacks
- Other approaches such as high performance OS dataplane architectures
Submissions must describe original, previously unpublished, complete research, not currently under review by another conference or journal.
Papers must be submitted electronically. The length of papers must be no more than 6 pages, including tables and figures, (in two-column, 10-point format) but excluding references, following the provided ACM LaTeX style file https://github.com/scyue/ccp-sigcomm18. The cover page must contain the name and affiliation of author(s) for single-blind peer reviewing by the program committee. Each submission will receive at least three independent blind reviews from the TPC. At least one of the authors of every accepted paper must register and present their work at the workshop.
Please submit your paper via http://sigcomm18kbnets.hotcrp.com/
Authors Take Note
The official publication date is the date the proceedings are made available in the ACM Digital Library. This date may be up to TWO WEEKS prior to the first day of the conference. The official publication date affects the deadline for any patent filings related to published work.
Attendance of the workshop is by open registration and subject to the same registration fees and rules as all the other SIGCOMM 2018 workshops. The registrants of the workshop may freely attend any workshop on the same day.
For the final paper to be published, please refer to Camera-ready instructions for workshops.
August 20, 2018
List of organization details
Program available online
June 10, 2018
April 30, 2018
April 01, 2018
Paper submission deadline
- Workshop Chairs
KAIST, South Korea
- Program Committee Members
D. K. Panda
KAIST, South Korea