Ertza Warraich
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ertza Warraich has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Muhammad Shahbaz
University of Michigan, Ann Arbor, USA
2 shared publications - Ali Imran
Department of Computer Science and Engineering, University of Michigan, Ann Arbor, MI, USA
1 shared publication Annus Zulfiqar
Department of Computer Science and Engineering, University of Michigan, Ann Arbor, MI, USA
1 shared publication- Annus Zulfiqar
University of Michigan, Ann Arbor, USA
1 shared publication - Marilyn Rego
University of Michigan, Ann Arbor, Michigan, USA
1 shared publication - Regan McDonald
University of Michigan, Ann Arbor, MI, USA
1 shared publication
Latest Publications
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 2033-2039
Towards Network-Efficient Cross-Regional Inference via Learned Activation Compression
Large transformer models are increasingly deployed across geographically distributed GPU clusters due to capacity, cost, and locality constraints. When inference is partitioned across sites, intermediate activations must be transmitted over wide area network (WAN) links at each partition boundary, introducing significant communication overhead. We present Feather, a system that reduces this overhead by compressing intermediate activations before transmission and reconstructing them before downstream layers resume execution. Feather learns a compact representation of activation tensors using a lightweight neural codec trained with a reconstruction objective while keeping the original model frozen. Across encoder and decoder transformer models, Feather achieves up to 48× activation compression while maintaining accuracy close to the baseline model. Under representative WAN conditions (e.g., 10 Gbps bandwidth and a 10 ms RTT), this reduction yields up to 4.96× improvement in end-to-end latency, consistently outperforming existing compression schemes, including linear autoencoder, PCA, and SVD.
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 2033-2039
Open AccessTowards Network-Efficient Cross-Regional Inference via Learned Activation Compression
Large transformer models are increasingly deployed across geographically distributed GPU clusters due to capacity, cost, and locality constraints. When inference is partitioned across sites, intermediate activations must be transmitted over wide area network (WAN) links at each partition boundary, introducing significant communication overhead. We present Feather, a system that reduces this overhead by compressing intermediate activations before transmission and reconstructing them before downstream layers resume execution. Feather learns a compact representation of activation tensors using a lightweight neural codec trained with a reconstruction objective while keeping the original model frozen. Across encoder and decoder transformer models, Feather achieves up to 48× activation compression while maintaining accuracy close to the baseline model. Under representative WAN conditions (e.g., 10 Gbps bandwidth and a 10 ms RTT), this reduction yields up to 4.96× improvement in end-to-end latency, consistently outperforming existing compression schemes, including linear autoencoder, PCA, and SVD.
- 22 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (2) , 393-396
Reimagining RDMA Through the Lens of ML
As distributed machine learning (ML) workloads scale to thousands of GPUs connected by ultra-high-speed interconnects, tail latency in collective communication has emerged as a primary bottleneck. Prior RDMA designs, like RoCE, IRN, and SRNIC, enforce strict reliability and in-order delivery, relying on retransmissions and packet sequencing to ensure correctness. While effective for general-purpose workloads, these mechanisms introduce complexity and latency that scale poorly, where even rare packet losses or delays can consistently degrade system performance. We introduce Celeris, a domain-specific RDMA transport that revisits traditional reliability guarantees based on ML's tolerance for lost or partial data. Celeris removes retransmissions and in-order delivery from the RDMA NIC, enabling best-effort transport that exploits the robustness of ML workloads. It retains congestion control (e.g., DCQCN) and manages communication with software-level mechanisms such as adaptive timeouts and data prioritization, while shifting loss recovery to the ML pipeline (e.g., using the Hadamard Transform). Early results show that Celeris reduces 99th-percentile latency by up to 2.3×, cuts BRAM usage by 67%, and nearly doubles NIC resilience to faults—delivering a resilient, scalable transport tailored for ML at cluster scale.
- 22 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (2) , 393-396
Reimagining RDMA Through the Lens of ML
As distributed machine learning (ML) workloads scale to thousands of GPUs connected by ultra-high-speed interconnects, tail latency in collective communication has emerged as a primary bottleneck. Prior RDMA designs, like RoCE, IRN, and SRNIC, enforce strict reliability and in-order delivery, relying on retransmissions and packet sequencing to ensure correctness. While effective for general-purpose workloads, these mechanisms introduce complexity and latency that scale poorly, where even rare packet losses or delays can consistently degrade system performance. We introduce Celeris, a domain-specific RDMA transport that revisits traditional reliability guarantees based on ML's tolerance for lost or partial data. Celeris removes retransmissions and in-order delivery from the RDMA NIC, enabling best-effort transport that exploits the robustness of ML workloads. It retains congestion control (e.g., DCQCN) and manages communication with software-level mechanisms such as adaptive timeouts and data prioritization, while shifting loss recovery to the ML pipeline (e.g., using the Hadamard Transform). Early results show that Celeris reduces 99th-percentile latency by up to 2.3×, cuts BRAM usage by 67%, and nearly doubles NIC resilience to faults—delivering a resilient, scalable transport tailored for ML at cluster scale.