After some time away from cloud computing (due to my new focus on conservation research), I served on the ACM SOCC 2024 program committee. It was quite interesting seeing the massive shift away from "traditional" cloud research (such as consensus protocols) towards many submissions aimed at accelerating machine learning workloads.

I also had a paper accepted there on decentralised scheduling, thanks to my former PhD student Smita Vijayakumar and her hard work on Murmuration!

References

• Vijayakumar et al (2024). Scheduling for Reduced Tail Task Latencies in Highly Utilized Datacenters. Association for Computing Machinery. 10.1145/3698038.3698522