RuleTris: Minimizing Rule Update Latency for TCAM-Based SDN Switches

Integrated Optical Content Addressable Memories (CAM) and Optical Random Access Memories (RAM) for Ultra-Fast Address Look-Up Operations

“…This can be of utter importance with the emergence of SDN and OpenFlow architectures that enforce a dynamic network operation with frequent updates of network topologies and multiple real time changes in the RIB-list [25,26]. Although the SDN controllers are increasingly optimized for swift policy updates, the T-CAM tables remain yet unoptimized for fast updates [27,28], which may trigger hundreds to thousands of table entry-moves and write-memory operations [26]. Measurements on the timings of such AL-table updates have revealed a few hundreds of ms-long response times [25][26][27][28], as AL-updates need to be organized in electronic T-CAM tables with sequential time-multiplexed memory accesses through a memory-bus of a limited bandwidth.…”

“…Although the SDN controllers are increasingly optimized for swift policy updates, the T-CAM tables remain yet unoptimized for fast updates [27,28], which may trigger hundreds to thousands of table entry-moves and write-memory operations [26]. Measurements on the timings of such AL-table updates have revealed a few hundreds of ms-long response times [25][26][27][28], as AL-updates need to be organized in electronic T-CAM tables with sequential time-multiplexed memory accesses through a memory-bus of a limited bandwidth. In this case, the optical nature of the proposed T-CAM may facilitate wavelength multiplexed memory access schemes in order to perform multiple simultaneous Write operations at 10 Gb/s, either when updating the prefix-list of the optical CAM-table or the outgoing ports stored in the optical RAM table, providing manifold improvements in the AL memory throughput.…”

“…Internet routers are experiencing 100 updates per second and potentially reaching upwards of 1000 per second [26], during which AL operations are stalled and the router remains idle, considerably limiting the performance. This, in-turn, requires fast Write operations to the CAM-table, with short latencies [25][26][27][28], to enable fast re-configurability of the network and rapid updating of the AL table.…”

Fast Network Configuration in Software Defined Networking

“…It is widely recognized that flow configuration, i.e., the configuration of flow forwarding rules at the SDN switches, is critical for SDN operation. Configuration updates due to changed flows must be performed consistently and quickly to avoid congestion [1], [2], [3], delays [4], loops and policy violations [3], [5], [6]. In particular, the need to reestablish disrupted flows caused by failing links as the result of a failure, congestion or attack on the network infrastructure, motivates the requirement to perform fast network reconfiguration.…”

“…The time required for deploying a given flow configuration is dominated by the time to insert/update flow rules in the ternary content addressable memory (TCAM) of each involved switch. Previous works such as [4], [7], [8], [9] report different per-rule update times ranging from 10ms up to 400ms depending on different models. The deployment time of network-wide flow rules on a data-center sized topology is not negligible since rule updates in a single switch must be performed sequentially to ensure consistency of rules [4], [5], [6], while multiple switches can be updated in parallel [5], [6].…”