Orion: Google's {Software-Defined} Networking Control Plane
We present Orion, a distributed Software-Defined Networking platform deployed globally in
Google's datacenter (Jupiter) and Wide Area (B4) networks. Orion was designed around a …
Google's datacenter (Jupiter) and Wide Area (B4) networks. Orion was designed around a …
Programmable calendar queues for high-speed packet scheduling
Packet schedulers traditionally focus on the prioritized transmission of packets. Scheduling
is often realized through coarse-grained queue-level priorities, as in today's switches, or …
is often realized through coarse-grained queue-level priorities, as in today's switches, or …
Aequitas: Admission control for performance-critical rpcs in datacenters
With the increasing popularity of disaggregated storage and microservice architectures, high
fan-out and fan-in Remote Procedure Calls (RPCs) now generate most of the traffic in …
fan-out and fan-in Remote Procedure Calls (RPCs) now generate most of the traffic in …
Congestion control in machine learning clusters
This paper argues that fair-sharing, the holy grail of congestion control algorithms for
decades, is not necessarily a desirable property in Machine Learning (ML) training clusters …
decades, is not necessarily a desirable property in Machine Learning (ML) training clusters …
[PDF][PDF] Deepweave: Accelerating job completion time with deep reinforcement learning-based coflow scheduling
P Sun, Z Guo, J Wang, J Li, J Lan, Y Hu - Proceedings of the Twenty-Ninth …, 2021 - ijcai.org
To improve the processing efficiency of jobs in distributed computing, the concept of coflow
is proposed. A coflow is a collection of flows that are semantically correlated in a multi-stage …
is proposed. A coflow is a collection of flows that are semantically correlated in a multi-stage …
Is advance knowledge of flow sizes a plausible assumption?
Recent research has proposed several packet, flow, and coflow scheduling methods that
could substantially improve data center network performance. Most of this work assumes …
could substantially improve data center network performance. Most of this work assumes …
dcPIM: Near-optimal proactive datacenter transport
Datacenter Parallel Iterative Matching (dcPIM) is a proactive data-center transport design
that simultaneously achieves near-optimal tail latency for short flows and near-optimal …
that simultaneously achieves near-optimal tail latency for short flows and near-optimal …
Packet order matters! improving application performance by deliberately delaying packets
Data centers increasingly deploy commodity servers with high-speed network interfaces to
enable low-latency communication. However, achieving low latency at high data rates …
enable low-latency communication. However, achieving low latency at high data rates …
On the hardness and inapproximability of virtual network embeddings
Many resource allocation problems in the cloud can be described as a basic Virtual Network
Embedding Problem (VNEP): the problem of finding a mapping of a request graph …
Embedding Problem (VNEP): the problem of finding a mapping of a request graph …
Leveraging service meshes as a new network layer
As modern cloud services have scaled out, applications have moved from relatively
monolithic designs to highly modularized fleets of microservices that communicate among …
monolithic designs to highly modularized fleets of microservices that communicate among …