End-Host Driven Troubleshooting Architecture for Software-Defined Networking

Abstract

Abstract-The high variability in traffic demands, the advanced networking services at various layers (e.g., loadbalancers), and the steady penetration of SDN technology and virtualization make the crucial network troubleshooting tasks ever more challenging over multi-tenant environments. Service degradation is first realized by the users and, as being the only one having visibility to many relevant information (e.g., connection details) required for accurate and timely problem resolution, the infrastructure layer is often forced upon continuous monitoring resulting in wasteful resource management, not to mention the long time frames. In this paper, we propose an End-host-Driven Troubleshooting architecture (EDT), where users are able to share the application-specific connection details with the infrastructure to accelerate the identification of root causes of performance degradation, and to avoid the need for always-on, resource-intensive, and networkwide monitoring. Utilizing EDT, we provide some essential tools for real end-to-end trace routing (PTR), identifying packet losses, and carry out hop-by-hop latency measurements (HEL). In contrast to existing proposals, PTR traces the practical production traffic without the need of crafted probe packets by means of careful tagging mechanisms and additional ephemeral capturing flow rules. Besides involving negligible data plane deterioration, in certain cases PTR can drastically reduce the time needed to find a traversed path compared to existing solutions. Finally, by means of individual network functions, HEL measures the latency of each link along the found path without involving the controller into the calculation, hence resulting in significant reduction of control plane overhead.

Reliable Control and Data Planes for Softwarized Networks

“…This insight can be used to adapt the fabric as the demand changes, migrate applications or virtual machines (VMs) to improve reliability [12], as well as improve policies to meet with the customer's Service Level Agreements (SLA) [10,11]. Moreover, telemetry data can be exploited in synergy with other logs to identify potential anomalies, security threats, and also for debugging [9,22]. For example, the Simple Network Management Protocol (SNMP) includes a widely standardised and adopted implementation of telemetry under both pull and push models.…”

Policy Injection

“…3). Attendees will also be engaged in discussions of, and be exposed to potential work-in-progress mitigation techniques and their trade-offs (e.g., joint troubleshooting techniques by tenants and provider [2], improved heuristics in OVS, flow cache-less softswitches [4]). Accompanying material can be found at https:// github.com/ cslev/ ovsdos.…”

Transition to SDN is HARMLESS: Hybrid Architecture for Migrating Legacy Ethernet Switches to SDN

“…Despite the fact that many large corporations (Google [7], Microsoft [8], Amazon [9]) and telcos (Orange, Verizon, Deutsche Telekom [10]) have already gained significant foothold in SDN, smaller businesses, campus network operators, and service providers seem reluctant to adopt it en masse [11]- [13]. In fact, enterprises without substantial in-house expertise and select IT staff ("organizations that aren't called Google" [2]) face significant business, economic, and technical deployment barriers, since migration to SDN requires a nontrivial amount of forward planning, an extensive investigation of vendor offerings and device options, and a fairly radical change in the mental model, producing a typical chicken and egg problem [1]- [6], [14].…”