Jupiter evolving: transforming google's datacenter network via optical circuit switches and software-defined networking
L Poutievski, O Mashayekhi, J Ong, A Singh… - Proceedings of the …, 2022 - dl.acm.org
We present a decade of evolution and production experience with Jupiter datacenter
network fabrics. In this period Jupiter has delivered 5x higher speed and capacity, 30 …
network fabrics. In this period Jupiter has delivered 5x higher speed and capacity, 30 …
Sundial: Fault-tolerant clock synchronization for datacenters
Clock synchronization is critical for many datacenter applications such as distributed
transactional databases, consistent snapshots, and network telemetry. As applications have …
transactional databases, consistent snapshots, and network telemetry. As applications have …
A throughput-centric view of the performance of datacenter topologies
While prior work has explored many proposed datacenter designs, only two designs, Clos-
based and expander-based, are generally considered practical because they can scale …
based and expander-based, are generally considered practical because they can scale …
Poseidon: Efficient, Robust, and Practical Datacenter {CC} via Deployable {INT}
The difficulty in gaining visibility into the fine-timescale hop-level congestion state of
networks has been a key challenge faced by congestion control (CC) protocols for decades …
networks has been a key challenge faced by congestion control (CC) protocols for decades …
Realizing RotorNet: Toward Practical Microsecond Scale Optical Networking
WM Mellette, A Forencich, R Athapathu… - Proceedings of the …, 2024 - dl.acm.org
We describe our experience building and deploying a demand-oblivious optically-switched
network based on the RotorNet and Opera architectures. We detail the design, manufacture …
network based on the RotorNet and Opera architectures. We detail the design, manufacture …
Performance trade-offs in reconfigurable networks for HPC
Designing efficient interconnects to support high-bandwidth and low-latency communication
is critical toward realizing high performance computing (HPC) and data center (DC) systems …
is critical toward realizing high performance computing (HPC) and data center (DC) systems …
Understanding lifecycle management complexity of datacenter topologies
Most recent datacenter topology designs have focused on performance properties such as
latency and throughput. In this paper, we explore a new dimension, life cycle management …
latency and throughput. In this paper, we explore a new dimension, life cycle management …
Gemini: Practical reconfigurable datacenter networks with topology and traffic engineering
To reduce cost, datacenter network operators are exploring blocking network designs. An
example of such a design is a" spine-free" form of a Fat-Tree, in which pods directly connect …
example of such a design is a" spine-free" form of a Fat-Tree, in which pods directly connect …
Experiences with modeling network topologies at multiple levels of abstraction
Network management is becoming increasingly automated, and automation depends on
detailed, explicit representations of data about the state of a network and about an operator's …
detailed, explicit representations of data about the state of a network and about an operator's …
Enabling quasi-static reconfigurable networks with robust topology engineering
Many optical circuit switched data center networks (DCN) have been proposed in the last
decade to attain higher capacity and topology reconfigurability, though commercial adoption …
decade to attain higher capacity and topology reconfigurability, though commercial adoption …