I am a senior principal research manager in the data systems group at Microsoft Research. I am interested in storage, key-value stores, streaming analytics, query processing, and distributed processing, and big data analytics for cloud and edge applications. My work on stream processing first shipped commercially in 2010 with Microsoft SQL Server, as the StreamInsight engine. I led the design and development of Trill, a streaming analytics engine that is widely used at Microsoft, for example, as part of the public-facing Azure Stream Analytics service.
Currently, I work on the SimpleStore research project, which has two components:
On the compute side, I lead the FASTER project which focuses on optimizing storage for compute and OLTP-style workloads, including serverless. Our artifacts include a high-performance concurrent key-value store and persistent log that supports larger-than-memory data. Our current focus is on distributed systems such as Shadowfax and remote FASTER, and serverless frameworks such as Netherite. We have also been exploring asynchronous recoverability, via general solutions such as CPR and DPR.
On the analytics side, I lead research work on simplifying and optimizing storage for analytics workloads, via proposals such as data layout (in the qd-tree project and its follow-up, MTO) and caching (in the Crystal project).
Learn more about my projects and check out my list of publications. I have served on the organizing and program committees of top database and systems conferences. Every summer, I enjoy working with students during their internships at Microsoft Research.
- Our paper on pushing computation to remote memory appeared at CIDR 2022:
- Q. Zhang, P. Bernstein, D. Berger, B. Chandramouli, V. Liu, B. T. Loo. CompuCache: Remote Computable Caching using Spot VMs. CIDR, 2022. [pdf]
- Our paper on remote caching is published in PVLDB 2022:
- Q. Zhang, P. Bernstein, D. Berger, B. Chandramouli. Redy: Remote Dynamic Memory Cache. PVLDB, 15(4), 2022. [pdf]
- I published three research papers at VLDB 2021:
- D. Durner, B. Chandramouli, Y. Li. Crystal: A Unified Cache Storage System for Analytical Databases. PVLDB, 14(11), 2021. [pdf]
- C. Kulkarni, B. Chandramouli, R. Stutsman. Achieving High Throughput and Elasticity in a Larger-than-Memory Store. PVLDB, 14(8), 2021. [pdf]
- W. Cai, P. A. Bernstein, W. Wu, B. Chandramouli. Optimization of Threshold Functions over Streams. PVLDB, 14(6), 2021. [pdf]
- Three research papers appeared at SIGMOD 2021:
- T. Li, B. Chandramouli, J. Faleiro, S. Madden, D. Kossmann. Asynchronous Prefix Recoverability for Fast Distributed Stores. SIGMOD 2021. [pdf]
- J. Ding, U. F. Minhas, B. Chandramouli, C. Wang, Y. Li, Y. Li, D. Kossmann, J. Gehrke, T. Kraska. Instance-Optimized Data Layouts for Cloud Analytics Workloads. SIGMOD 2021. [pdf]
- A. Arasu, B. Chandramouli, J. Gehrke, E. Ghosh, D. Kossmann, J. Protzenko, R. Ramamurthy, T. Ramananandro, A. Rastogi, S. Shetty, N. Swamy, A. van Renen, M. Xu. FastVer: Making Data Integrity a Commodity. SIGMOD 2021. [pdf]
- A paper I led at MSR, with my intern and others, on learning data layouts in storage for big data analytics appeared SIGMOD 2020. We propose a data structure called a qd-tree (built using deep RL) to layout data blocks in a workload-guided manner. Easy to integrate into DB as well!
- We have designed and built an updatable and adaptive learned index called ALEX. The paper appeared at SIGMOD 2020, and you can find a copy below.
- Jialin Ding et al. ALEX: An Updatable Adaptive Learned Index. SIGMOD 2020. [pdf]
- I introduced the SimpleStore umbrella research project at HPTS. See my slides here.
- A research paper and demo on FishStore, for fast ingestion, storage, and indexing of raw data, appeared at SIGMOD 2019 and VLDB 2019.
- Badrish Chandramouli, Dong Xie, Yinan Li, Donald Kossmann. FishStore: Fast Ingestion and Indexing of Raw Data. VLDB 2019, Los Angeles, California, USA, August 2019 (demo). [pdf]
- Dong Xie, Badrish Chandramouli, Yinan Li, Donald Kossmann. FishStore: Faster Ingestion with Subset Hashing. SIGMOD 2019, Amsterdam, Netherlands, June 2019. [pdf]
- We have a fresh take on the age-old database recovery problem. Learn about Concurrent Prefix Recovery (CPR) in our research paper at SIGMOD 2019.
- Guna Prasaad, Badrish Chandramouli, Donald Kossmann. Concurrent Prefix Recovery: Performing CPR on a Database. SIGMOD 2019, Amsterdam, Netherlands, June 2019. [pdf]
- A short paper on our open-source system, CRA (Common Runtime for Applications), appeared at ICDE 2019. CRA is the backbone distributed runtime behind Quill and Ambrosia.
My Twitter Feed
Tweets by badrishc