Minlan Yu
Title: Application customized networking
Abstract: As applications become highly distributed, the need for high-performance communication and coordination between servers becomes critical. Unfortunately, today's network abstractions and network stacks significantly limit performance optimization due to their generic for diverse applications: Traditional network abstractions focus on individual connections, missing the potential benefits of coordinating groups of connections in distributed systems. Efforts on network-layer performance optimization sometimes lead to more computation at applications. Layers of network stacks in the kernel, which are essential for security and resource sharing, introduce significant overhead.
In this talk, I advocate for application-customized networking, which empowers applications to specify their own communication needs and optimizations in the network layer. This approach enables applications to coordinate across connections, perform end-to-end optimizations, and reduce kernel overhead, all while retaining the security and resource-sharing benefits of the kernel. I will give two examples of application-customized networking: First, for distributed transaction systems, we introduce DINT, which leverages eBPF to support frequent-path transaction operations directly in the kernel. DINT achieves up to 2.6× higher throughput than using a DPDK-based kernel-bypass stack, with only a maximum of 10% increase in average unloaded latency. Second, for distributed deep learning, we introduce THC, a bidirectional tensor homomorphic compression framework that enables the direct aggregation of compressed values and thus eliminates the computational overheads of (de)compression at the parameter server. THC reaches target accuracy up to 1.47× faster compared with state-of-the-art systems.
Bio: Minlan Yu is a Gordon McKay professor at Harvard School of Engineering and Applied Science. She’s the assistant director of the SRC/DARPA JUMP 2.0 ACE Center for Evolvable Computing. She received her B.A. in computer science and mathematics from Peking University and her M.A. and PhD in computer science from Princeton University. She has actively collaborated with companies such as Google, AT&T, Microsoft, Facebook, and Intel. Her research interests include data networking, distributed systems, enterprise and data center networks, and software-defined networking. She received the ACM-W rising star award, NSF CAREER award, and ACM SIGCOMM doctoral dissertation award. She served as PC co-chair for SIGCOMM, NSDI, HotNets, and several other conferences and workshops.
Carlos Cordeiro
Title: “Beyond Speed: Infusing Determinism and Intelligence into the Future of Wi-Fi”
Abstract: Wi-Fi, a cornerstone of global wireless connectivity, is estimated to carry anywhere between 50% to 80% of the world's Internet traffic volume. Recent strides in Wi-Fi technology, exemplified by Wi-Fi 6 and Wi-Fi 7, showcase remarkable achievements in meeting escalating data demands through innovations like OFDMA, wider 320 MHz channels, 4K QAM modulation, and multi-link operation, enabling aggregate peak data rates exceeding 30 Gbps. While these advancements are commendable, the future of Wi-Fi extends beyond raw speed. Anticipating the next decade, this presentation centers on the imperative for Wi-Fi to evolve into a reliable, deterministic, and intelligent network foundation. This evolution is essential to cater to emerging applications such as contextual computing, environmental sensing, multi-device experiences, gesture recognition, AR/VR, ambient and industrial IoT, and gaming. As such, in this talk, we navigate through Wi-Fi's historical journey, and its current state, and illuminate the major technological trends shaping the next decade. Emphasis is placed on the development of key technologies required to bring determinism and intelligence to Wi-Fi, thus addressing the diverse and complex needs of the future.
Bio: Dr. Carlos Cordeiro is an Intel Fellow and serves as the wireless CTO in Intel’s client computing group. Carlos leads Intel’s global wireless connectivity standards and ecosystem team and is responsible for defining Intel’s next generation wireless connectivity technology strategy and development, ecosystem engagements, and regulatory planning. Carlos has had a leading role in the technology development, standardization, and productization of various generations of Wi-Fi and Bluetooth, having developed major technology innovations that are found in billions of wireless devices. He is an IEEE Fellow and has served as the chairman of the Board of Directors of the Wi-Fi Alliance, the organization that defines Wi-Fi.
Due to his contributions to wireless communications, Carlos received several awards including the Wireless Broadband Alliance’s prestigious CTO of the Year Award in 2022, the 2017 IEEE Standards Medallion, the Intel Inventor of the Year Award in 2016, the IEEE Outstanding Engineer Award in 2011, and the IEEE New Face of Engineering Award in 2007. He is the co-author of two textbooks on wireless published in 2006 and 2011, has published over 130 papers in the wireless area alone, and holds over 450 patents. He is the editor-in-chief of the IEEE Communications Standards Magazine and has served as Editor of various journals including the IEEE Transactions on Mobile Computing, the IEEE Journal on Selected Areas in Communications, the IEEE Transactions on Wireless Communications, the IEEE Wireless Communication Letters and the ACM Mobile Computing and Communications Review journal.