Dejan Kostić
where else to find them
no profile links on file
what they talk about
where they keep going back
- NSDI2 editions2020–2023
all 6 in the filters
on a stage with
all 18 in the filters
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| Queue-Mem: Energy-Efficient Hardware Storage for Advanced Network Function Acceleration | NSDI '26 | 2026-05-04 |
Queue-Mem stores packet payloads in the switch shared packet buffer by pausing/resuming port queues with a batch-based queue control that maps five-tuple… | ||
| Toward Automated, Contamination-free Dafny Benchmark Generation | Dafny 2026 | 2026-01-11 |
TACoDafny automates converting task descriptions (first 200 LeetCode problems) into Dafny programs plus contracts by prompting LMs to produce implementations… | ||
| A High-Speed Stateful Packet Processing Approach for Tbps Programmable Switches | NSDI 2023 | 2023-04-17 |
RIBOSOME splits packets at a Tofino ASIC, sends only headers to a FastClick NF on one 100G NIC and stores payloads on shared servers via RDMA, achieving 300… | ||
| RDMA is Turing complete, we just did not know it yet! | 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22) | 2022-04-04 |
Claims RDMA is Turing complete by running self-modifying RDMA chains on commodity NICs: pre-posted RDMA_RECEIVE/READ/WRITE operations get mutated at runtime… | ||
| Packet Order Matters! Improving Application Performance by Deliberately Delaying Packets | 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22) | 2022-04-04 |
Builds Reframer, a user-space reordering layer that classifies packets into per-flow queues using a hash table plus a flush linked list, briefly buffers them… | ||
| Reexamining Direct Cache Access to Optimize I/O Intensive Applications for Multi | USENIX ATC '20 | 2020-07-15 |
Microbenchmarks on two servers with 100 Gbps NICs show DDIO (Intel DDIO) causes about a 30% increase in 99th‑percentile latency when aggregate receive rate… | ||
| A High-Speed Load-Balancer Design with Guaranteed Per-Connection-Consistency | NSDI 2020 | 2020-02-25 |
Cheetah embeds a per-connection server decision into a packet cookie (implemented as a stateless cookie in TCP timestamp or QUIC/IPv6 alternatives) so packets… | ||
| Metron: NFV Service Chains at the True Speed of the Underlying Hardware | NSDI '18 | 2018-04-09 |
Metron is an NFV platform that splits service chains into stateless hardware offloads (OpenFlow switches, Mellanox/DPDK NICs) and stateful software (fastClick)… | ||
| nothing matches | ||