Summary of 2024

This blog post is about summarizing various highlights and events that happened in 2024. Since July I have joined IBM Research Lab in Zurich as Senior Research Scientist. I continue to collaborate with my PhD students in VU Amsterdam. Here are some of our scientific and academic outputs:

Highlights

• 2024 Best paper award at the 15th ACM/SPEC International Conference on Performance Engineering (ICPE ‘24) for the paper BFQ, Multiqueue-Deadline, or Kyber? Performance Characterization of Linux Storage Schedulers in the NVMe Era, London, UK.
• Invited talk at the fourth Workshop on Challenges and Opportunities of Efficient and Performant Storage Systems (CHEOPS’24) https://cheops-workshop.github.io/2024.html co-located with EuroSys 2024, Anthens, Greece.
• 2024 CCGrid paper on Reviving Storage Systems Education in the 21st Century - An experience report is nominated for the best paper award, https://2024.ccgrid-conference.org/awards/.

Read more »

How to ask for a recommendation

“Ask whether I can write you a good letter, not whether I’m willing to write one.” – Geoff Kuenning

So you have just finished a project or thesis or course with me, and want me to write a recommendation for you? Perhaps for a job or higher education application. Here are quick guidelines (for both BSc and MSc works):

Read more »

Reflection on organizing CompSys 2021

CompSys is the Dutch local flagship conference (https://www.compsys.science/) where Dutch PhD students and researchers showcase their research. It is held annually, and this year I organized the conference. This year also, it was the first time that the conference was organized online. For me it was a unique experience to organize and run the conference.

Read more »

My experience with 6 months of online teaching

I have been teaching online from September 2020 till March 2021. During this time, I started two new courses - Advanced Network Programming and Storage Systems, and continued running a paper reading seminar Advanced Topics in Distributed Systems. Due to their nature (seminar, lecture with programming homework), all three courses entailed a very different nature of interaction with students. Let me summarize my experience with these classes and my general experience and struggles.

Read more »

How to use VU DAS-5 system

DAS-5 or DAS (The Distributed ASCI Supercomputer 5 is VU’s local cluster installation. In this guide I will tell you how to use it for your project work.

Read more »

How to setup SoftiWARP on Ubuntu 18.04 and run an RDMA server-client example

A brief write-up about how to build, and run SoftiWARP RDMA device in an Ubuntu 18.04 installation, and an RDMA server-client code example.

https://github.com/animeshtrivedi/blog/blob/master/post/2019-06-26-siw.md

Investigating performance of Apache Arrow (Java) in-memory (part 3)

A brief write-up about investigating the read performance for Java and C++ for Apache Arrow format in-memory for integer schema type https://github.com/animeshtrivedi/blog/blob/master/post/2018-11-22-arrow-cpp.md

Investigating performance of Apache Arrow (Java) in-memory (part 2)

A brief write-up about investigating the read performance of Apache Arrow format in-memory for integers schema type https://github.com/animeshtrivedi/blog/blob/master/post/2018-10-09-arrow-int.md

Investigating performance of Apache Arrow (Java) in-memory (part 1)

A brief write-up about investigating the read performance of Apache Arrow format in-memory for byte[] schema type https://github.com/animeshtrivedi/blog/blob/master/post/2018-10-03-arrow-binary.md

Investigating the RoCE (v2) protocol

A brief write-up about packet and operation rates on a RoCE v2 link https://github.com/animeshtrivedi/blog/blob/master/post/2018-09-13-RoCEv2-header.md.

SQL Performance: Part 1 - Input File Formats

Cross-posting from http://crail.incubator.apache.org/blog/2018/08/sql-p1.html

How to read and write Apache Arrow files in Java

https://github.com/animeshtrivedi/blog/blob/master/post/2017-12-26-arrow.md.

How to run Crail File System locally using netty

In this blog post I am going to show how to run Crail file system (CrailFS) from www.crail.io (github) locally on your laptop using its netty binding.

Read more »

How does Apache Spark read a parquet file

In this post I will try to explain what happens when Apache Spark tries to read a parquet file. Apache Parquet is a popular columnar storage format which stores its data as a bunch of files. Typically these files are stored on HDFS. In a seprate post I will explain more details about the internals of Parquet, but for here we focus on what happens when you call

Read more »