新闻与深度文章
Welcome to Research Focus, a series of blog posts that highlights notable publications, events, code/datasets, new hires and other milestones from across the research community at Microsoft. Time-series forecasting is a technique used to predict future values based on previously…
In this issue: New research on appropriate reliance on generative AI; Power management opportunities for LLMs in the cloud; LLMLingua-2 improves task-agnostic prompt compression; Enhancing COMET to embrace under-resourced African languages:
| Rodrigo Fonseca, Pedro Las-Casas, Alok Kumbhare, 和 Ricardo Bianchini
Cloud Intelligence/AIOps research from Microsoft could help organizations autonomously manage the entire cloud platform. Find out how.
| Daniel S. Berger, Marisa You, Celine Irvene, Mark Jung, Tyler Narmore, Jacob Shapiro, Luke Marshall, Savyasachi Samal, Preetha Subbarayalu, Ashish Raniwala, Brijesh Warrier, 和 Ricardo Bianchini
Managing server failures at the scale of a cloud platform is challenging. The Hyrax fail-in-place approach reduces the need for immediate repairs and creates a path toward lowering water consumption and carbon emissions in cloud datacenters.
| Konstantina Mellou, Marco Molinaro, 和 Rudy Zhou
Efficiency is vital in the face of escalating demand for cloud resources. And efficient power management strategies address the bottleneck of power availability in datacenters. Learn how we optimize power allocation to support sustainable resource usage.
奖项 | USENIX OSDI
Daniel Berger receives OSDI 2023 Best Paper Award
Daniel Berger’s paper, Enso: A Streaming Interface for NIC-Application Communication, was selected as a best paper by the OSDI’23 organizers for its quality and potential impact. Enso was also awarded for its open source artifacts.
| Rodrigo Fonseca
EuroSys 2023 (opens in new tab) is the premier systems conference in Europe, and 2023 marks its 18th edition. Sponsored by ACM SIGOPS Europe and hosted May 8 to May 12, the conference covers a wide range of topics, including…
Our paper, Pond: CXL-Based Memory Pooling Systems for Cloud Platforms, was selected as a distinguished paper by the ASPLOS’23 organizers. Congrats to Daniel S. Berger and Pantea Zardoshti .
奖项 | ACM – SOSP
Daniel Berger receives SOSP 2021 Best Paper Award
Daniel Berger’s paper, Kangaroo: Caching Billions of Tiny Objects on Flash, was selected as a best paper by the SOSP’21 organizers for its quality and potential impact.