I'm a member of the research staff at IBM research Zurich. My research interests are in distributed systems, networking and operating systems. I graduated with a PhD from ETH Zurich in 2008 and spent two years (2008-2010) as a Postdoc at Microsoft Research Silicon Valley.
The general theme of my work is to explore how modern networking and storage hardware can be exploited in distributed systems. Over the last years, I've been working on Crail, a fast distributed data store designed from ground up for fast storage (DRAM, NVMe, PCM) and networking hardware (100Gb/s RDMA, NVMf). Crail is built upon principles of user-level I/O and primarily targets fast sharing of ephemeral data in distributed data processing workloads (Spark, Tensforflow, serverless workloads, etc.).
Currently, I'm working on a new storage platform for efficient ML training on disaggregated storage.
Earlier, I developed DiSNI, a zero-copy RDMA-based network stack for the JVM, DaRPC, a ultra-low latency RPC library also for the JVM, and jVerbs, an RDMA-based network stack and precurser of DiSNI, which is part of the IBM JDK since May 2014.
- Paper "Unification of Temporary Storage in the NodeKernel Architecture" was accepted to Usenix ATC'19!
- April 2019: Slides from my recent Oreilly Strata talk on Crail are up!
- March 2019: New blog post on deployment options for tiered storage disaggregation
- March 2019: New blog post on shuffle disaggregation using RDMA and NVMf
- October 2018: Paper on ephemeral storage for serverless computing accepted to OSDI'18
- Spark Summit'18 talk on serverless machine learning using Crail
- Two papers accepted at Usenix ATC 2018
- January 2018: Crail is now an Apache Incubator project: crail.apache.org
- January 2017: Crail website is up: www.crail.io, moved to crail.apache.org
- October 2016: our (ongoing) work on the Crail I/O architecture is available at github.com/apache/incubator-crail
- July 2016: DiSNI library available at github.com/zrlio/disni
- July 2016: DaRPC library available at github.com/zrlio/darpc
- Pocket: Ephemeral Storage for Serverless Analytics, Ana Klimovic, Yawen Wang, Christos Kozyrakis, Patrick Stuedi, Jonas Pfefferle, and Animesh Trivedi, OSDI'18, Carlsbad, CA, Octoboer 2018
- Understanding Ephemeral Storage for Serverless Analytics. Ana Klimovic, Yawen Wang, Christos Kozyrakis, Patrick Stuedi, Jonas Pfefferle, and Animesh Trivedi, USENIX ATC, Boston, USA, July 2018.
- Albis: High-Performance File Format for Big Data Systems. Animesh Trivedi, Patrick Stuedi, Jonas Pfefferle, Adrian Schuepbach, and Bernard Metzler, USENIX ATC, Boston, USA, July 2018.
- Crail: A High-Performance I/O Architecture for Distributed Data Processing: Patrick Stuedi, Animesh Trivedi, Jonas Pfefferle, Bernard Metzler, Ioannis Koltsidas, Radu Stoica, Nikolas Ioannou, IEEE Data Bulletin
- FlashNet: Flash/Network Stack Co-Design: Animesh Trivedi, Nikolas Ioannou, Bernard Metzler, Patrick Stuedi, Jonas Pfefferle, Ioannis Koltsidas, Kornilios Kourtis, Thomas R. Gross, Systor 2017, Haifa, Israel
- On the [Ir]Relevance of Network Performance for Data Processing: Animesh Trivedi, Patrick Stuedi, Radu Stoica, Bernard Metzler, Ioannis Koltsidas, Nikolas Ioannou, HotCloud 2016
- A Hybrid I/O Virtualization Framework for RDMA-Capable Network Interfaces: Jonas Pfefferle, Patrick Stuedi, Animesh Trivedi, Bernard Metzler, Ioannis Koltsidas, Thomas Gross, ACM VEE'15, March 2015
- DaRPC: Data Center RPC: Patrick Stuedi, Animesh Trivedi, Bernard Metzler, Jonas Pfefferle, ACM SOCC'14, Seattle, WA, November 2014
- JVerbs: Ultra-Low Latency for Data Center Applications: Patrick Stuedi, Bernard Metzler, Animesh Trivedi, ACM SOCC'13, San Jose, CA, October 2013
- Unified High-Performance I/O, One Stack to Rule Them All: Animesh Trivedi, Patrick Stuedi, Bernard Metzler, Roman Pletka, Blake Fitch, Thomas Gross, HotOS'13, Santa Ana Pueblo, NM, May 2013
- Pasture: Secure Offline Data Access using Commodity Trusted Hardware: Rama Kotla, Tom Rodeheffer, Indranjit Roy, Patrick Stuedi, Benjamin Wester, OSDI'12, Hollywood, CA, October 2012
- Wimpy Nodes with 10GbE: Leveraging One-Sided Operations in Soft-RDMA to boost Memcached: Patrick Stuedi, Animesh Trivedi, Bernard Metzler, (Short Paper), Usenix'12, Boston, MA, June 2012
- Data Processing at the Speed of 100 Gbps using Apache Crail, Oreilly Strata, San Francisco, March 2019
- COMPASS Talk about Apache Crail, ETH Zurich, September 2019 [slides]
- Data Processing at the Speed of 100 Gbps using Apache Crail, VMware Research, Palo Alto, August 2018
- Serverless Machine Learning on Modern Hardware using Apache Crail, Spark Summit'18, San Francisco, June 2018
- Running Spark on High-Performance Hardware using RDMA and NVMe Flash, Spark Summit'17, San Francisco, June 2018
- Wimpy Nodes with 10Gb: Leveraging One-Side Operations in Soft-RDMA to boost Memcached, Usenix ATC'12, Boston, June 2012
I'm co-teaching the following courses at ETH Zurich.
- Advanced Computer Networks FS 2018
- Advanced Computer Networks FS 2017
- Advanced Computer Networks FS 2016
- Advanced Computer Networks FS 2015
- Advanced Computer Networks FS 2014
- Advanced Computer Networks FS 2013
- Advanced Computer Networks FS 2012
- Advanced Computer Networks FS 2011
- Usenix ATC 2019
- ASPLOS 2019 (external review comittee)
- SOCC 2018
- Usenix ATC 2018
- SOCC 2017
- Systor 2017
- ICDCS 2014
- ICDCS 2011
- ICDCS 2010