] My first paper got accepted in APSys 2017.
NrOS: Effective Replication and Sharing in an Operating System
Ankit Bhardwaj, Chinmay Kulkarni, Reto Achermann, Irina Calciu, Sanidhya Kashyap, Ryan Stutsman,
Amy Tai, and Gerd Zellweger
15th USENIX Symposium on Operating Systems Design and Implementation (OSDI '21) - To appear
Adaptive Placement for In-memory Storage Functions
Ankit Bhardwaj, Chinmay Kulkarni, and Ryan Stutsman
2020 USENIX Annual Technical Conference (ATC '20)
[ Paper ] [ Slides ] [ Talk ] [ Code ]
On the Impact of Isolation Costs on Locality-aware Cloud Scheduling
Ankit Bhardwaj, Meghana Gupta, and Ryan Stutsman
12th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud '20)
[ Paper ] [ Slides ] [ Talk
A Preliminary Performance Model for Optimizing Software Packet Processing
Ankit Bhardwaj, Atul Shree, Bhargav Reddy V, and Sorav Bansal
8th ACM SIGOPS Asia-Pacific Workshop on Systems (APSys '17)
[ Paper ] [ Poster ] [ Slides ] [ Talk ]
COP: Compiler Optimizations to Reduce Memory Stalls for Network Pipelines
Written in P4
Shailja Pandey, Ankit Bhardwaj, Anmol Panda and Sorav Bansal
Poster in 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI '18
[ Abstract ] [ Poster ]
I am broadly interested in Distributed Systems, Operating Systems, and Cloud Computing.
Bespin - is an new
operating system written from scratch in Rust with the goal to explore the kernel replication on each
NUMA node and sharing the replica among cores such that it can scale for a multi-socket machine. It uses
operation logging to synchonize the replicated kernel across sockets.
Splinter - is a
multi-tenant in-memory key-value store that allows pushing storage functions at runtime. Traditional
techniques to provide isolation among tenants are too costly and can't be used for a μs-scale system.
So, our group is looking for novel ways to solve the problem.
Bacus - At IIT Delhi, my research focus was on compiler-based optimizations for software-based
packet processing. We worked on optimizing the compilers so that DSLs(like P4) can
be used more effectively. I was a part of Vertecs Paper Reading group; details can be found here.
vLAB - At IIT Bombay, I worked on vLab: Managing
and Provisioning VMs for Labs, where the aim was to run a large number of virtual machines on
commodity hardware. I designed and implemented the system from scratch, and also analyzed the impact of
various parameters on VM boot time when images were stored on distributed object store setup over a
cluster of commodity machines.
Fall 2020, Distributed Systems, University of Utah