ACM SIGCOMM 2024 Workshop on Hot Topics in Optical Technologies and Applications in Networking (HotOptics 2024)

Invited Speakers:

Prof. Elaine Wong (University of Melbourne)

Title: Challenges and Emerging Approaches to Support Heterogenous Services in Next-Generation Access Networks

Description: By 2030, an average broadband user will be connected to up to 15 devices - a majority of which will have extended reality and multisensory capabilities, thus allowing communications to be fully experiential and immersive. Future access and end user networks must therefore be able to support a range of heterogenous services across the cyber physical world continuum. In this talk, we will first motivate the continued push to advance access networks to support new services in new industries and rationalize the use of machine intelligence in these networks. We then highlight the challenges that are especially amplified due to the service dynamicity and heterogeneity, data scarcity, and computation-resource constraints of access networks. We discuss emerging approaches such as concept drift and explainability, that are being explored to address such challenges.

Dr. Reza Nejabati (Cisco)

Title: Quantum Leap: Cisco's Vision for a Groundbreaking Quantum Network

Description: In this talk, I will explore Cisco's innovative approach to quantum networking and its real-world applications. Specifically, I will review Cisco's achievements and outline the roadmap for developing a practical and effective quantum network. This network will support distributed quantum computing and sensing while also enhancing quantum-safe communication.

Prof. Keren Bergman (Columbia University)

Title: Petascale Photonic Connectivity for Energy-Efficient Computing

Description: High performance systems are increasingly bottlenecked by the energy and communications costs of interconnection networks. Leveraging silicon photonics connectivity to alleviate the bandwidth/energy challenges, a new architectural approach enables flexible connectivity tailored to accelerate distributed ML applications.

Prof. George Papen (University of California San Diego)

Title: Practical Challenges for Future Optical Networks

Description: Optical networking is now being used in production systems for both traditional data center networks and machine learning applications. This talk presents some of the key practical challenges for future optical networks and describes how they were addressed in developing a practical microsecond realization of RotorNet.

Program

Call for Papers

Optical technologies represent a vibrant community of science and engineering innovations, from Nobel-prize-winning breakthroughs (e.g., the invention of optical fibers, optical lasers, etc.) to hundreds of billion-dollar industries (e.g., in telecommunications, cloud computing, semiconductors, and silicon photonics, etc.). Today, the global Internet and digital infrastructures are built upon optical networks and systems. In particular, the area of computer communications and networking, where the SIGCOMM community is thriving, has significantly benefited from the advancement of optical technologies in the past several decades. For example, the advent of long-haul wavelength-division multiplexing (WDM) optical fibers in the 1970s paved the way for upper-layer innovations including TCP/IP and wide-area networks. The development of integrated photonics since the 2000s has led to the mass deployment of optical transceivers in modern data center networks. The recent strategic deployment of optical circuit switching (OCS) by industry giants like Google underscores the potential and readiness for large-scale reconfigurable optical interconnects for emerging machine-learning workloads.

The SIGCOMM community is committed to advancing the technologies of computer networks and networked systems. Over the past several years, as the slowdown of Moore's law is affecting the entire electronic industry, we have witnessed an emerging trend of adopting optical hardware to sustain the ever-growing traffic demands, ranging from optical wide-area networks to optical circuit-switched data center networks. Most recently, new more radical research efforts have emerged in the community to explore novel optical hardware like silicon photonic Tbps optical I/O, smart optical transceivers capable of photonic computing, and optically reconfigurable TPU to build high-performance networked systems to accelerate machine learning workloads. This line of work blends emerging optical hardware into the existing cloud networked systems at the crossroads of multiple research areas, e.g., silicon photonics, fiber optics, FPGA, programmable data plane, scheduling, synchronization, congestion control, and network optimization, which requires cross-disciplinary insights and knowledge. Despite the promising and exciting outlook, we do hear comments and feedback from members of the SIGCOMM community on many occasions about the high barrier of entry into optical networks research due to the lack of optics-related background knowledge, open-source hardware kits and software tools, etc.

This workshop aims to embrace the synergistic history between the optics community and networking community and explore the next exciting opportunities that lie at the intersection of these two areas. To facilitate a fruitful discussion on this interdisciplinary topic, we will invite lead experts from both areas to talk about state-of-the-art advances in this area. We will also call for technical submissions from the community to build a successful workshop.

Topics of Interest

The HotOptics workshop aims to provide a premier forum for researchers, practitioners, and professionals to discuss and advance the latest developments in emerging optical technologies and their applications in networking. This workshop seeks contributions that explore innovative ideas, challenge existing paradigms, and expand the boundaries of research in optical networking technologies.

In particular, the HotOptics workshop seeks to blend novel physical-layer optical technologies with upper-layer computer networking protocols and applications to unlock new synergies. We encourage both bottom-up research that develops new optical devices and systems to enable new features and a top-down approach that integrates these novel features into advanced applications and services. Topics of Interest include, but are not limited to:

Bottom-up Approach: Optical Devices and Systems for Advanced Optical Networks

• Bottom-up Approach: Optical Devices and Systems for Advanced Optical Networks

  • On-chip optical interconnect networks
  • Inter-chip optical interconnects and optical I/O
  • Commercial or Engineering prototypes (optical modulators, optical detectors, optical transceivers, optical amplifiers, etc.)
  • Photonic network elements and hardware (e.g., wavelength-selective switches, ROADMs, etc)
  • Hollow-core, micro-structured, multi-core, few-mode and multi-mode optical fibers
  • Fiber sensing devices and systems in optical networks
  • Silicon photonics devices and systems
  • Real-time implementation of advanced modulation formats, FEC, and DSP in ASICs and FPGAs in communication networks
  • Machine-learning-based techniques for optical devices and system design
  • Novel free-space optical communication and networking devices
  • Satellite optical communication links
  • VCSEL modules and Systems for communication networks
  • Radio-over-Fiber (RoF) systems and networks
  • Intelligent Surfaces for Optical Wireless and THz Communication

• Top-down Approach: New Applications and Services enabled by Optical Networks and Systems

  • Time synchronization over optical networks
  • Anchored Modeling (digital-twin) of optical network systems
  • Cross-layer Optical-IP networks control and resource orchestration
  • Optical network planning, optimization, and techno-economic studies
  • Quantum-secured optical networks
  • Machine learning applications in optical networks
  • Optical technologies for emerging machine learning systems
  • Security and privacy issues in optical networks
  • Energy efficiency and sustainability in optical networking
  • Open/disaggregated optical network architecture
  • Advanced computing and storage systems using optical technologies
  • Reconfigurable optical network architecture and systems
  • Operational aspects of optical network applications, including SDN, NFV, Network Management, Monitoring, and Analytics (AIOps)

Submission Instructions

We welcome submissions of two types of contributions:

• Short papers:

  • This type of submission presents original, unpublished research work not currently under review by other conferences or journals, and short position papers or vision papers that identify new research problems, advocate for new research methodology, or offer critiques and debate of existing work in the field. These submissions do not need to be supported by full evaluations - preliminary and proof-of-concept demonstrations are sufficient. Any of the following categories of short papers are welcome. (1) Position papers that present a future vision or critiques of optical networking research. (2) Self-sustained work with solid technical contribution and evaluation, but the scope is smaller than a 12-page full paper. (3) Early-stage work with an initial research idea and preliminary results, to be extended to a full paper later.
  • The paper should not exceed 6 pages (10 point font, 12 point leading, 7 inch by 9.25 inch text block), including figures, and tables. References and appendix are allowed to go beyond the page limit. The submitted research paper will be peer-reviewed by the program committee in a double-blind way, and those accepted papers will be presented during the workshop, and published in the final workshop proceedings at ACM Digital Library.
  • The goal of this type of research paper is to provide a venue for researchers to publish and receive timely feedback on their ongoing work or not-yet-mature ideas so that eventually the final 12-page full paper can land in top systems and optics venues such as SIGCOMM, NSDI, OSDI, SOSP, ASPLOS, as well as journals papers like Nature and Science portfolios, Journal of Optical Communications and Networking (JOCN), Journal of Lightwave Technology (JLT), Optics Express, etc.

• Technical abstracts:

  • This type of submission presents a summary of early-stage or ongoing work seeking community feedback. We also welcome submissions on short abstracts or highlights of previously published work.
  • The submitted technical abstract should not exceed 3 pages (10 point font, 12 point leading, 7 inch by 9.25 inch text block), including figures, tables, references, and appendix. The submitted abstracts will be peer-reviewed by the program committee in a single-blind way, and those accepted abstracts will be presented during the workshop. Technical abstracts will not appear in the final workshop proceedings, and not be considered as publications.
  • The goal of this type of technical abstract is to provide a venue for researchers to advertise their previously published results to reach a broader audience and seek follow-up collaborations that often involve interdisciplinary expertise.

Prospective authors are expected to submit papers written in English using the standard ACM template for conference papers (https://github.com/scyue/latex-sigcomm18). If you are using LaTeX, you may make use of this template for ACM conference proceedings. With the older versions of this template, you must add “10pt” to the documentclass command to meet the submission requirements. The current template sets 10pt by default. (Unlike the official template, it only includes examples for conference proceedings.)

Submissions must be made via the HotOptics website (https://hotoptics24.hotcrp.com/). At least one author from each accepted submission must attend the workshop to present and discuss their work.

If you have any questions, please contact workshop chairs Jesse E. Simsarian (jesse.simsarian@nokia-bell-labs.com), Massimo Tornatore (massimo.tornatore@polimi.it), Yiting Xia (yxia@mpi-inf.mpg.de), and Zhizhen Zhong (zhizhenz@mit.edu).

• Submissions deadline

  May 28, 2024 (firm)
• Acceptance notification

  June 10, 2024
• Camera-ready deadline

  June 14, 2024
• Workshop

  Aug 4, 2024

Paper Conflicts and Peer Review

As part of the paper submission process, authors will be required to provide paper metadata which includes title and abstract, author names, affiliations, contact email addresses, topics matching the subject matter of the paper, and conflicts with program committee (PC) members.

Following the same criteria as SIGCOMM 2024, we define conflict of interest with a PC member using the following principles:

• You are currently employed at the same institution, have been previously employed at the same institution within the last 12 months, or are going to begin employment at the same institution.
• You have a professional partnership as follows:
  • Past or present association as advisor or advisee.
  • Collaboration on a project, publication, or grant proposal within the past 2 years (i.e., 2022 or later).

The workshop chairs and members will review conflicts to ensure the integrity of the reviewing process, adding conflicts where necessary and sanity-checking cases where conflicts do not appear justified. If there is no basis for PC conflicts provided by authors, those conflicts will be removed. Improperly identifying PC members as a conflict to avoid individual reviewers may lead to your paper being rejected. If you have concerns, please contact the PC chairs.

The 6-page research papers will go through a double-blind peer review by the program committee, where the identities of the authors are withheld from the reviewers. As an author, you are required to make a good-faith effort to preserve the anonymity of your submission, while at the same time allowing the reader to fully grasp the context of related past work, including your own. Common sense and careful writing will go a long way toward preserving anonymity. Minimally, please take the following steps when preparing your submission:

• Remove the names and affiliations of authors from the title page.
• Remove acknowledgment of identifying names and funding sources
• Use care in naming your files. Source file names (e.g., “John-Doe.dvi”) are often embedded in the final output as readily accessible comments
• Use care in referring to related work, particularly your own. Do not omit references to provide anonymity, as this leaves the reviewer unable to grasp the context. Instead, reference your past work in the third person, just as you would any other piece of related work by another author
• Do not embed pointers to external sources (e.g., public GitHub repositories) that leak author identity or affiliation

The 3-page technical abstracts will go through a single-blind peer review by the program committee, where the identities of the authors and the affiliation are visible to the reviewers, while the identities of reviewers are withheld from the authors. This is because the 3-page technical abstracts allow authors to submit highlights of their previously published work, and will not appear as workshop publications in the ACM Digital Library.

Travel Support

The HotOptics workshop aims to bring researchers from both optics/photonics and computer systems and networks communities together and promote interdisciplinary work at the crossroads of these areas. We plan to provide travel grant support for authors/presenters of accepted papers/presentations to travel to Sydney for a lively discussion and debate. Please stay tuned for further travel grant application information.

Organizers