11 citations · 24 across the 8 of their papers we have counts for
9 papers
PCM: Programmable Congestion Management for Terabit RDMA NICs
Mikhail Khalilov, Timo Schneider, Tommaso Bonato +10
Datacenter congestion control (CC) and load balancing (LB) evolve rapidly as workload demands change, but deployment depends on RDMA hardware and transport interfaces. Emerging pro…
In-Network Collective Operations: Game Changer or Challenge for AI Workloads?
Torsten Hoefler, Mikhail Khalilov, Josiah Clark +9
This paper summarizes the opportunities of in-network collective operations (INC) for accelerated collective operations in AI workloads. We provide sufficient detail to make this i…
EDAN: Towards Understanding Memory Parallelism and Latency Sensitivity in HPC
Siyuan Shen, Mikhail Khalilov, Lukas Gianinazzi +6
Resource disaggregation is a promising technique for improving the efficiency of large-scale computing systems. However, this comes at the cost of increased memory access latency d…
Hazel: Secure and Efficient Disaggregated Storage
Marcin Chrapek, Meni Orenbach, Ahmad Atamli +4
Disaggregated storage with NVMe-over-Fabrics (NVMe-oF) has emerged as the standard solution in modern supercomputers and data center clusters, achieving superior performance, resou…
SDR-RDMA: Software-Defined Reliability Architecture for Planetary Scale RDMA Communication
Mikhail Khalilov, Siyuan Shen, Marcin Chrapek +16
RDMA is vital for efficient distributed training across datacenters, but millisecond-scale latencies complicate the design of its reliability layer. We show that depending on long-…
Understanding Data Movement in Tightly Coupled Heterogeneous Systems: A Case Study with the Grace Hopper Superchip
Luigi Fusco, Mikhail Khalilov, Marcin Chrapek +3
Heterogeneous supercomputers have become the standard in HPC. GPUs in particular have dominated the accelerator landscape, offering unprecedented performance in parallel workloads…
Network-Offloaded Bandwidth-Optimal Broadcast and Allgather for Distributed AI
Mikhail Khalilov, Salvatore Di Girolamo, Marcin Chrapek +3
In the Fully Sharded Data Parallel (FSDP) training pipeline, collective operations can be interleaved to maximize the communication/computation overlap. In this scenario, outstandi…
REPS: Recycled Entropy Packet Spraying for Adaptive Load Balancing and Failure Mitigation
Tommaso Bonato, Abdul Kabbani, Ahmad Ghalayini +7
Next-generation datacenters require highly efficient network load balancing to manage the growing scale of artificial intelligence (AI) training and general datacenter traffic. How…
OSMOSIS: Enabling Multi-Tenancy in Datacenter SmartNICs
Mikhail Khalilov, Marcin Chrapek, Siyuan Shen +7
Multi-tenancy is essential for unleashing SmartNIC's potential in datacenters. Our systematic analysis in this work shows that existing on-path SmartNICs have resource multiplexing…