&lt;italic&gt;NetStar&lt;/italic&gt;: A Future/Promise Framework for Asynchronous Network Functions

Duan, Jingpu; Yi, Xiaodong; Wang, Junjie; Wu, Chuan; Le, Franck

doi:10.1109/jsac.2019.2894303

Cited by 8 publications

(11 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We can see that the NFs both record a transmission rate of about 10 Mpps when we allocate 24 cores to FlexState. These results are in line with the performance recorded by the NFs using other state management systems [4], [14]. We can conclude that FlexState, similarly to existing state management solutions, allows NFs to approach line-rate packet processing.…”

Section: Does Flexstateenable Nfs To Approach Line Rate?supporting

confidence: 88%

“…The Bro intrusion detection system (IDS) [13] uses timers to handle cases in which the response of a DNS query is not received within certain time budget [14]. To manage the state of Bro, a network developer could initially adopt Split/Merge because it supports timers (Table 1).…”

Section: Examplementioning

confidence: 99%

“…FlexStateleverages RSS to distribute the flows across the available cores and partitioning to avoid inter-core con-tention, and thus improve the performance and the scalability of the system. In our implementation, we use Seastar [26], a framework that has been used successfully in other related work [14]. Seastar distributes the flows across the available cores by configuring the NIC to apply RSS and linking each hardware queue of the NIC to a different core.…”

Section: B State Managementmentioning

confidence: 99%

“…We configure Pktgen on node1 to send traffic to node2 saturating the 10 Gbps link, i.e.,14.88 Mpps [14]. We run testpmd on node2 and we configure it to send the received traffic back to node1.…”

Section: Does the Testbed Support Line Rate?mentioning

confidence: 99%

“…TIMERS Some NFs use timers to carry out their tasks. For example, when a new flow arrives, a malware detector performs a query to a registry for malware signatures, and it arms a timer to be able to react in case no reply is provided [14]. For this reason, some state management systems offer explicit support for storing timers and notifying the NF in case of expiration [1], [4].…”

Section: Locksmentioning

confidence: 99%

See 4 more Smart Citations

FlexState: Flexible State Management of Network Functions

et al. 2021

View full text Add to dashboard Cite

Network function (NF) developers have traditionally prioritized performance when creating new packet processing capabilities. This was usually driven by a market demand for highly available solutions with differentiating features running at line rate, even at the expense of flexibility and tightly-coupled monolithic designs. Today, however, the market advantage is achieved by providing more features in shorter development cycles and quickly deploying them in different operating environments. In fact, network operators are increasingly adopting continuous software delivery practices as well as new architectural styles (e.g., microservices) to decouple functionality and accelerate development. A key challenge in revisiting NF design is state management, which is usually highly optimized for a deployment by carefully selecting the underlying data store. Therefore, migrating to a data store that suits a different use case is time-consuming as it requires code refactoring and adaptation to new application programming interfaces, APIs. As a result, refactoring NF software for different environments can take up to months, reducing the pace at which new features and upgrades can be deployed in production networks. In this paper, we demonstrate experimentally that it is feasible to introduce an abstraction layer to decouple NF state management from the data store adopted while still approaching line-rate performance. We present FlexState, a state management system that exposes data store functionality as configuration options, which reduces code refactoring efforts. Experiments show that FlexState achieves significant flexibility in optimizing the state management, and accelerates deployment on new scenarios while preserving performance and scalability.

show abstract

Section: Does Flexstateenable Nfs To Approach Line Rate?supporting

confidence: 88%

Section: Examplementioning

confidence: 99%

Section: B State Managementmentioning

confidence: 99%

Section: Does the Testbed Support Line Rate?mentioning

confidence: 99%

Section: Locksmentioning

confidence: 99%

See 3 more Smart Citations