Despite software-defined networking's proven benefits, there remains a significant reluctance in adopting it. Among the issues that hamper SDN's adoption, two issues stand out: reliability and fault tolerance. At the heart of these issues is a set of fate-sharing relationships: the first between the SDN control applications and controllers, wherein the crash of the former induces a crash of the latter, thereby affecting the controller's availability; and, the second between the SDN-Apps and the network, wherein the failure of the former violates network safety, e.g., network-loops, or network availability, e.g., black holes.In this paper, we argue for a redesign of the controller architecture centering around a set of abstractions to eliminate these fate-sharing relationships and thus improve the controller's availability. We present a prototype implementation of a framework, called LegoSDN, that embodies our abstractions, and we demonstrate the benefits of our abstractions by evaluating LegoSDN on an emulated network with five real SDN-Apps. Our evaluations show that LegoSDN can recover failed SDN-Apps 3× faster than controller reboots while simultaneously preventing policy violations.
Network failures are inevitable. Interfaces go down, devices crash and resources become exhausted. It is the responsibility of the control software to provide reliable services on top of unreliable components and throughout unpredictable events. Guaranteeing the correctness of the controller under all types of failures is therefore essential for network operations. Yet, this is also an almost impossible task due to the complexity of the control software, the underlying network, and the lack of precision in simulation tools.Instead, we argue that testing network control software should follow in the footsteps of large scale distributed systems, such as those of Netflix or Google, which deliberately induce live failures in their production environments during working hours, and analyze how their control software reacts.In this paper, we describe Armageddon, a framework for introducing sustainable and systematic chaos in networks. When we cause failures, we do so without violating some operator-specified network invariants (e.g., end-to-end connectivity). The injected failures also guarantee some notion of coverage. If the controller can sustain all of the failures, then it can be considered resilient with a high degree of confidence. We describe efficient algorithms to compute failure scenarios and implemented them in a prototype. Applied to real-world networks, our algorithms a coverage of 80% of the links within only three iterations of failures.
Many geo-distributed services at web-scale companies still rely on databases (DBs) primarily optimized for single-site performance. At AT&T this is exemplified by services in the network control plane that rely on third-party software that uses DBs like MariaDB and PostgreSQL, which do not provide strict serializability across sites without a significant performance impact. Moreover, it is often impractical for these services to re-purpose their code to use newer DBs optimized for geo-distribution. In this paper, a novel drop-in solution for DB clustering across sites called Metric is presented that can be used by services without changing a single line of code. Metric leverages the single-site performance of an existing service's DB and combines it with a cross-site clustering solution based on an entry-consistent redo log that is specifically tailored for geo-distribution. Detailed correctness arguments are presented and extensive evaluations with various benchmarks show that Metric outperforms other solutions for the access patterns in our production use-cases where service replicas access different tables on different sites. In particular, Metric achieves up to 56% less latency and 5.2x higher throughput than MariaDB and PostgreSQL clustering, and up to 90% less latency and 26x higher throughput than CockroachDB and TiDB, systems that are designed to support geo-distribution.
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.