I specialize in the security and performance of solid-state drives (SSD). My research focuses on developing methods and techniques to improve both the efficiency and security of SSDs by utilizing DRAM caches and incorporating machine learning models.

1. Ransomware Detection by Leveraging DRAM Cache of SSD:  Accepted: MEMSYS2024 

• By utilising the DRAM cache of the SSD, we effectively detect ransomware. In addition to aiding ransomware detection, the DRAM cache also contributes to reducing the overhead associated with data recovery.

2. DRAM Cache-Assisted Garbage Collection: Submitted : ASPDAC2025

• We propose techniques that leverage the onboard DRAM cache of SSDs to assist in garbage collection by reducing the migration of cached pages from the DRAM cache. This approach helps in minimizing garbage collection blockage time.

3. Machine Learning-Based Dead Page Detection: TBD by: Noverber 2024

• We utilize machine learning models such as random forests, SVM, and decision tree to identify dead pages entering the SSD’s DRAM cache. These models help decide whether an incoming I/O request (read/write to a page) should be cached in the DRAM or written directly to the flash memory. This work is currently in progress.

4. Erasure-Based Secure Deletion Schemes: TBD by: Jan2025

• This work focuses on reducing the overhead associated with erasure-based secure deletion schemes in SSDs. In this work we focus on reducing the count of page relocation on each block erasure. This work is in progress.