billions of packets It's hard to outperform CPUs, and you need a *large* market (e.g. GPUs) to pay to keep that speed over the long term. But putting data movement code outside the reach of the CPU makes the problem itself much easier, as you can eliminate huge amounts of protection overhead.

God bless reply guys

My #1 PhD student @butro successfully completed his PhD defense. Thanks to the committee (Jignesh Patel billions of packets Sam Madden). Matt's thesis is on accelerating databases with eBPF (telemetry, proxies, OLTP stores). You have 60 days to hire him. Expect fierce competition.

Which citation style is this? Dropping the title seems like a great way of saving space

I am thrilled to share that the thesis of my student, Udit Paul, co-advised with Elizabeth Belding, has been selected as the winner of this year's ACM SIGCOMM Doctoral Dissertation Award for his fundamental contributions to bridging the data divide.

This is my first Dagstuhl, and we are all busy pondering the past, present, and future state of Programmable Host Networking. Thanks, Gianni Antichi, Aurojit Panda, katerina Argyraki, and billions of packets, for the invitation and spearheading the discussions! 🙏

Y'all they've been ghosting me since 9AM

My phone is back! The social media team was much snappier than the main support line and the driver was lovely. Thank heavens.

Junior US Academics! Come learn about the junior faculty career path and engagement with the National Science Foundation! sites.google.com/view/nets-earl…

"Latency Guarantees for Caching with Delayed Hits" w/ Keerthana Gurushankar & Bernardo Subercaseaux 🐒 was accepted @ INFOCOM'25. Started from a project in billions of packets Nirav Atre 's networks class at CMU Computer Science Department ! tl;dr LRU is competitive even when inter-request time << storage latency