Validating datacenters at scale

Jayaraman, Karthick; Bjørner, Nikolaj; Padhye, Jitendra; Agrawal, Amar; Bhargava, Ashish; Bissonnette, Paul-Andre C; Foster, Shane; Helwer, Andrew; Kasten, Mark; Lee, Ivan; Namdhari, Anup; Niaz, Haseeb; Parkhi, Aniruddha; Pinnamraju, Hanukumar; Power, A. Kathryn; Raje, Neha Milind; Sharma, Parag

doi:10.1145/3341302.3342094

Cited by 43 publications

(33 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rigorous analysis proves the convergence and correctness of CPCheck. It also shows that the previous published Azure local verification [27] can be easily supported as a specialized case of verifying shortestpath reachability and fault-tolerance requirement in CPCheck.…”

Section: Introductionmentioning

confidence: 84%

“…Each entry in the table maps a disjoint packet space [29] to an action. The action includes modification of packet header as well as a group of next-hops [27]. it is annotated with a keyword ALL.…”

Section: Data Plane Modelmentioning

confidence: 99%

“…First, most verification tools rely on a centralized verification server to collect snapshots of the network states or configuration files from all of the network devices, which requires reliable connections between the server and network devices, resulting in a bottleneck and a single point of failure. Azure [27] proposed to have devices verify its forwarding behaviors using local contracts. However, they only support verifications of the shortest path reachability and fault tolerance property.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Carbide: Highly Reliable Networks Through Real-Time Multiple Control Plane Composition

Chen

Gökarslan

et al. 2020

Preprint

View full text Add to dashboard Cite

Achieving highly reliable networks is essential for network operators to ensure proper packet delivery in the event of software errors or hardware failures. Networks must ensure reachability and routing correctness, such as subnet isolation and waypoint traversal. Existing work in network verification relies on centralized computation at the cost of fault tolerance, while other approaches either build an over-engineered, complex control plane, or compose multiple control planes without providing any guarantee on correctness. This paper presents Carbide, a novel system to achieve high reliability in networks through distributed verification and multiple control plane composition. The core of Carbide is a simple, generic, efficient distributed verification framework that transforms a generic network verification problem to a reachability verification problem on a directed acyclic graph (DAG), and solves the latter via an efficient distributed verification protocol (DVprotocol). Equipped with verification results, Carbide allows the systematic composition of multiple control planes and realization of operator-specified consistency. Carbide is fully implemented. Extensive experiments show that (1) Carbide reduces downtime by 43% over the most reliable individual underlying control plane, while enforcing correctness requirements on all traffic; and (2) by systematically decomposing computation to devices and pruning unnecessary messaging between devices during verification, Carbide scales to a production data center network.

show abstract

Section: Introductionmentioning

confidence: 84%

Section: Data Plane Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Carbide: Highly Reliable Networks Through Real-Time Multiple Control Plane Composition

Chen

Gökarslan

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…These approaches scale to networks with a couple of 100k forwarding rules and some support incremental updates [4], [18] which allow reverification times in the range of milliseconds. Further improvements have been made by dividing snapshots into smaller but closed portions [21], [22] which are easier to analyze and allow massive parallelization [23]. Additional improvements can be made by limiting on IP routed networks instead of an arbitrary amount of header fields [24].…”

Section: Overview Of Related Verification Toolsmentioning

confidence: 99%

Continuous Verification of Network Security Compliance

Lorenz¹,

Clemens²,

Schrötter³

et al. 2021

Preprint

View full text Add to dashboard Cite

Continuous verification of network security compliance is an accepted need. Especially, the analysis of stateful packet filters plays a central role for network security in practice. But the few existing tools which support the analysis of stateful packet filters show runtimes in the order of minutes to hours making them unsuitable for continuous compliance verification.<br>In this work, we address these challenges and present a solution which is based on the application of formal methods. First, we introduce the formal language FPL that enables a high-level human-understandable specification of the desired state of network security. Second, we demonstrate the instantiation of a compliance process using a verification framework that analyzes the configuration of complex networks and devices - including stateful firewalls - for compliance with FPL policies. Our evaluation results show the scalability of the presented approach for the well known Internet2 and Stanford benchmarks as well as for large firewall rule sets where it outscales state-of-the-art tools by a factor of over 41.

show abstract

“…This development is also not limited to academia. Large cloud providers, such as Alibaba [36], Amazon [2], and Microsoft [20], are developing and deploying network verification systems. However, as shown in Figure 1, each such tool today is a monolith, with its own model of the target functionality and its own analysis engine.…”

Section: Introductionmentioning

confidence: 99%

A General Framework for Compositional Network Modeling

Beckett

Mahajan

2020

Proceedings of the 19th ACM Workshop on Hot Topics in Networks

View full text Add to dashboard Cite

We advocate for an approach to network modeling and analysis based on a common intermediate language. Unlike today, where each tool builds a custom model and analysis engine for its target network functionality, we argue that network functionality should be expressed in a common language. This approach makes it easier to expand formal analysis to new functionality and analyze interactions between dependent functionalities (e.g., routing and packet filtering). We demonstrate the feasibility of this approach by developing an intermediate language called Zen and three different analyses for programs in that language. For representative data plane and control plane functionalities, we find that Zen reduces the modeling effort by an order of magnitude, while providing analysis performance that matches custom tools. CCS CONCEPTS • Networks → Network reliability; • Theory of computation → Program verification; • Software and its engineering → Model checking; Functional languages.

show abstract

Validating datacenters at scale

Cited by 43 publications

References 27 publications

Carbide: Highly Reliable Networks Through Real-Time Multiple Control Plane Composition

Carbide: Highly Reliable Networks Through Real-Time Multiple Control Plane Composition

Continuous Verification of Network Security Compliance

A General Framework for Compositional Network Modeling

Contact Info

Product

Resources

About