RECORD
PowerTCP: Pushing the Performance Limits of Datacenter Networks
Networking and Internet Architecture · Submitted 28 Dec 2021
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim We show analytically and empirically that PowerTCP can significantly outperform the state-of-the-art in both traditional datacenter topologies and emerging reconfigurable datacenters where frequent bandwidth changes make congestion control challenging.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Increasingly stringent throughput and latency requirements in datacenter networks demand fast and accurate congestion control. We observe that the reaction time and accuracy of existing datacenter congestion control schemes are inherently limited. They either rely only on explicit feedback about the network state (e.g., queue lengths in DCTCP) or only on variations of state (e.g., RTT gradient in TIMELY). To overcome these limitations, we propose a novel congestion control algorithm, PowerTCP, which achieves much more fine-grained congestion control by adapting to the bandwidth-window product (henceforth called power). PowerTCP leverages in-band network telemetry to react to changes in the network instantaneously without loss of throughput and while keeping queues short. Due to its fast reaction time, our algorithm is particularly well-suited for dynamic network environments and bursty traffic patterns. We show analytically and empirically that PowerTCP can significantly outperform the state-of-the-art in both traditional datacenter topologies and emerging reconfigurable datacenters where frequent bandwidth changes make congestion control challenging. In traditional datacenter networks, PowerTCP reduces tail flow completion times of short flows by 80% compared to DCQCN and TIMELY, and by 33% compared to HPCC even at 60% network load. In reconfigurable datacenters, PowerTCP achieves 85% circuit utilization without incurring additional latency and cuts tail latency by at least 2x compared to existing approaches.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Increasingly stringent throughput and latency requirements in datacenter networks demand fast and accurate congestion control. | background 1.00 | 0.05 | 0.02 | asserted 0.99 |
| s1 | We observe that the reaction time and accuracy of existing datacenter congestion control schemes are inherently limited. | gap 0.54 | 0.73 | 0.04 | asserted 1.00 |
| s2 | They either rely only on explicit feedback about the network state (e.g., queue lengths in DCTCP) or only on variations of state (e.g., RTT gradient in TIMELY). | background 0.62 | 0.26 | 0.02 | asserted 1.00 |
| s3 | To overcome these limitations, we propose a novel congestion control algorithm, PowerTCP, which achieves much more fine-grained congestion control by adapting to the bandwidth-window product (henceforth called power). | contribution 0.96 | 0.87 | 0.23 | asserted 1.00 |
| s4 | PowerTCP leverages in-band network telemetry to react to changes in the network instantaneously without loss of throughput and while keeping queues short. | method 0.81 | 0.75 | 0.13 | asserted 1.00 |
| s5 | Due to its fast reaction time, our algorithm is particularly well-suited for dynamic network environments and bursty traffic patterns. | interpretation 0.97 | 0.44 | 0.07 | asserted 0.93 |
| s6 | We show analytically and empirically that PowerTCP can significantly outperform the state-of-the-art in both traditional datacenter topologies and emerging reconfigurable datacenters where frequent bandwidth changes make congestion control challenging. | result 0.77 | 0.96 | 0.69 | asserted 1.00 |
| s7 | In traditional datacenter networks, PowerTCP reduces tail flow completion times of short flows by 80% compared to DCQCN and TIMELY, and by 33% compared to HPCC even at 60% network load. | result 1.00 | 0.96 | 0.19 | asserted 1.00 |
| s8 | In reconfigurable datacenters, PowerTCP achieves 85% circuit utilization without incurring additional latency and cuts tail latency by at least 2x compared to existing approaches. | result 1.00 | 0.95 | 0.23 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Increasingly stringent throughput and latency requirements in datacenter networks demand fast and accurate congestion control. We observe that the reaction time and accuracy of existing datacenter congestion control schemes are inherently limited. They either rely only on explicit feedback about the network state (e.g., queue lengths in DCTCP) or only on variations of state (e.g., RTT gradient in TIMELY). To overcome these limitations, we propose a novel congestion control algorithm, PowerTCP, which achieves much more fine-grained congestion control by adapting to the bandwidth-window product (henceforth called power). PowerTCP leverages in-band network telemetry to react to changes in the network instantaneously without loss of throughput and while keeping queues short. Due to its fast reaction time, our algorithm is particularly well-suited for dynamic network environments and bursty traffic patterns. We show analytically and empirically that PowerTCP can significantly outperform the state-of-the-art in both traditional datacenter topologies and emerging reconfigurable datacenters where frequent bandwidth changes make congestion control challenging. In traditional datacenter networks, PowerTCP reduces tail flow completion times of short flows by 80% compared to DCQCN and TIMELY, and by 33% compared to HPCC even at 60% network load. In reconfigurable datacenters, PowerTCP achieves 85% circuit utilization without incurring additional latency and cuts tail latency by at least 2x compared to existing approaches.
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.