Viraj is usually tinkering with hardware and software. Always looking towards working at the intersection of new, exciting and stressful.
His research focused on disaggregated database systems, tuning LSM-KVS, cutting edge hardware devices (SMR, ZNS), and integration of modern AI techniques in database systems.
Publications
Probe Less, Learn More: Characterizing and Correcting eBPF Instrumentation Overhead
ProTools Workshop @ SC 2026
·
To Appear
TellyTune: Exploring the Limits of LLM-Assisted Tuning for LSM-Based Key-Value Stores
ACM SYSTOR 2026
·
02 Sep 2026
·
doi:10.1145/3793230.3839381
ANCHOR: A Vision for Secure Persistent Key-Value Stores in Disaggregated Data Centers
ACM SeQureDB 2026
·
To Appear
·
Preprint (arXiv:2608.27819)
O3-LSM: Maximizing Disaggregated LSM Write Performance via Three-Layer Offloading
ACM SIGMOD 2026
·
18 May 2026
·
doi:10.1145/3802093
SHIELD: Encrypting Persistent Data of LSM-KVS from Monolithic to Disaggregated Storage
ACM SIGMOD 2025
·
17 Jun 2025
·
doi:10.1145/3725354
Can Modern LLMs Tune and Configure LSM-based Key-Value Stores?
ACM HotStorage 2024
·
08 Jul 2024
·
doi:10.1145/3655038.3665954
CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure
ACM SIGMOD 2024
·
29 May 2024
·
doi:10.1145/3654927