Modern distributed storage requires microsecond-scale tail latency, but the current coordinator-based quorum coordination causes a burdensome latency overhead. This paper presents Archon, a new quorum coordination architecture that supports low tail latency for microsecond-scale replicated storage. The key idea of Archon is to perform the quorum coordination in the network switch by leveraging the flexibility and capability of emerging programmable switch ASICs. Our in-network quorum coordination is based on the observation that the modern programmable switch provides nanosecond-scale processing delay and high flexibility simultaneously. To realize the idea, we design a custom switch data plane. We implement a Archon prototype on an Intel Tofino switch and conduct a series of testbed experiments. Our experimental results show that Archon can provide lower tail latency than the coordinator-based solution.
Gyuyeong KIM
Sungshin Women's University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Gyuyeong KIM, "Switch-Based Quorum Coordination for Low Tail Latency in Replicated Storage" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 11, pp. 1922-1925, November 2023, doi: 10.1587/transinf.2023EDL8038.
Abstract: Modern distributed storage requires microsecond-scale tail latency, but the current coordinator-based quorum coordination causes a burdensome latency overhead. This paper presents Archon, a new quorum coordination architecture that supports low tail latency for microsecond-scale replicated storage. The key idea of Archon is to perform the quorum coordination in the network switch by leveraging the flexibility and capability of emerging programmable switch ASICs. Our in-network quorum coordination is based on the observation that the modern programmable switch provides nanosecond-scale processing delay and high flexibility simultaneously. To realize the idea, we design a custom switch data plane. We implement a Archon prototype on an Intel Tofino switch and conduct a series of testbed experiments. Our experimental results show that Archon can provide lower tail latency than the coordinator-based solution.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023EDL8038/_p
Copy
@ARTICLE{e106-d_11_1922,
author={Gyuyeong KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Switch-Based Quorum Coordination for Low Tail Latency in Replicated Storage},
year={2023},
volume={E106-D},
number={11},
pages={1922-1925},
abstract={Modern distributed storage requires microsecond-scale tail latency, but the current coordinator-based quorum coordination causes a burdensome latency overhead. This paper presents Archon, a new quorum coordination architecture that supports low tail latency for microsecond-scale replicated storage. The key idea of Archon is to perform the quorum coordination in the network switch by leveraging the flexibility and capability of emerging programmable switch ASICs. Our in-network quorum coordination is based on the observation that the modern programmable switch provides nanosecond-scale processing delay and high flexibility simultaneously. To realize the idea, we design a custom switch data plane. We implement a Archon prototype on an Intel Tofino switch and conduct a series of testbed experiments. Our experimental results show that Archon can provide lower tail latency than the coordinator-based solution.},
keywords={},
doi={10.1587/transinf.2023EDL8038},
ISSN={1745-1361},
month={November},}
Copy
TY - JOUR
TI - Switch-Based Quorum Coordination for Low Tail Latency in Replicated Storage
T2 - IEICE TRANSACTIONS on Information
SP - 1922
EP - 1925
AU - Gyuyeong KIM
PY - 2023
DO - 10.1587/transinf.2023EDL8038
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2023
AB - Modern distributed storage requires microsecond-scale tail latency, but the current coordinator-based quorum coordination causes a burdensome latency overhead. This paper presents Archon, a new quorum coordination architecture that supports low tail latency for microsecond-scale replicated storage. The key idea of Archon is to perform the quorum coordination in the network switch by leveraging the flexibility and capability of emerging programmable switch ASICs. Our in-network quorum coordination is based on the observation that the modern programmable switch provides nanosecond-scale processing delay and high flexibility simultaneously. To realize the idea, we design a custom switch data plane. We implement a Archon prototype on an Intel Tofino switch and conduct a series of testbed experiments. Our experimental results show that Archon can provide lower tail latency than the coordinator-based solution.
ER -