报告人：K. K. Ramakrishnan 教授 University of California, Riverside
主持人：黎仁蔚
报告时间：2024年8月23日（周五）上午10:00-11:30
报告地点：九龙湖校区计算机楼513会议室
报告人简介：Dr. K. K. Ramakrishnan is Distinguished Professor of Computer Science and Engineering at the University of California, Riverside. Previously, he was a Distinguished Member of Technical Staff at AT&T Labs-Research. He joined AT&T Bell Labs in 1994 and was with AT&T Labs-Research since its inception in 1996. Prior to 1994, he was a Technical Director and Consulting Engineer in Networking at Digital Equipment Corporation. Between 2000 and 2002, he was at TeraOptic Networks, Inc., as Founder and Vice President.
Dr. Ramakrishnan is an ACM Fellow, IEEE Fellow, and an AT&T Fellow, recognized for his fundamental contributions to communication networks, congestion control, traffic management, VPN services, and a lasting impact on AT&T and the industry. His work on the DECbit congestion avoidance protocol received the ACM Sigcomm Test of Time Paper Award in 2006. K. K. received the ACM Sigcomm Lifetime Achievement Award in 2024. He has published over 300 papers and has over 180 patents issued in his name. K.K. has been on the editorial board of several journals and has served as the TPC Chair and General Chair for several networking conferences. K. K. received his MTech from the Indian Institute of Science (1978, recently recognized as one of IISc’s Distinguished Alumni), MS (1981) and Ph.D. (1983) in Computer Science from the University of Maryland, College Park, USA.
报告摘要：Communication networks are changing. They are becoming more and more “software-based”, especially with the use of Network Function Virtualization (NFV) to run network services in software, and networking virtualized components in the cloud. I will use a couple of our recent efforts to illustrate what we have learned.
Using our high-performance NFV platform, OpenNetVM we developed a high-performance, low-latency core for 5G cellular networks. Our core, L25GC+, re-architects the 5G core (5GC) network, and its processing, to reduce latency of control plane operations and their impact on the data plane. Exploiting shared memory, L25GC+ eliminates message serialization and HTTP processing overheads, while being 3GPP-standards compliant. L25GC+ reduces event completion time by ~50% for several control plane events and improves data packet latency (due to improved control plane communication) by ~2×, during paging and handover events. But we realize that truly achieving high performance requires us to also re-think the protocols we use in cellular networks, not just implement the same set of protocols on a competent system. Holistic solutions that exploit the use of flexible software platforms and adapt network protocols to eliminate unnecessary message exchanges can truly offer significant benefits.
A fast-growing sub-area of cloud computing is the use of Serverless Computing to simplify the development, deployment, and automated management of modular software functions, exploiting the microservices paradigm while promising efficient, low-cost compute capability for users. However, serverless computing has been enabled by integrating multiple cloud computing software components to rapidly offer its capabilities in the cloud while sacrificing the efficient networking of the virtualized components. We exploit event-driven shared memory processing to dramatically improve dataplane scalability, by eliminating data copies, and avoiding unnecessary protocol processing and serialization-deserialization overheads. We also find the use of the extended Berkeley Packet Filter (eBPF) enables the creation of true event-driven processing, especially to replace the typical heavyweight sidecar proxy used in serverless computing. Overall, we achieve an order of magnitude improvement in throughput and latency compared to Knative, while substantially reducing CPU usage, and mitigating the need for `cold-start'.
