DACON: a reconfigurable application-centric optical network for disaggregated data center infrastructures [Invited]

Abstract

To solve the issues of low resource utilization and performance bottleneck in current server-centric data center networks (DCNs), we propose and experimentally demonstrate a disaggregated application-centric optical network (DACON) for data center infrastructures based on hybrid optical switches. DACON achieves flexible provision and dynamic reconfiguration of hardware nodes exploiting the softwared-define networking (SDN) orchestration plane. A four-node SDN-enabled disaggregated prototype is implemented with a field-programmable-gate-array-based controller of hardware nodes and nanosecond optical switches, performing a minimal node-to-node network latency of 378.6 ns and zero packet loss. Based on the unmodified Linux kernel and two different applications (distributed computing and a Memcached database), the application runtime of the disaggregated prototype is investigated and compared with the server-centric architecture. Experimental results show that the disaggregated prototype performs better with Memcached database applications, achieving a 1.46 × faster runtime than the server-centric network at a memory node access ratio of 0.9. Based on the customized control plane orchestrator and dynamic resource reallocation, the node-to-node latency is reduced by 21% when CPU nodes access memory nodes. The scalability of DACON is then numerically assessed based on experimentally measured parameters. Results show that the intra-rack node-to-node latency is less than 404.8 ns with a 6144-node network and memory node access ratio of 0.9. Finally, the cost and power consumption are also studied and compared with current DCN architectures. Results indicate DACON saves 13.4% of the cost of an interconnect network compared with current disaggregated architecture and consumes up to 31.1% less power with respect to server-centric DCN architectures.

On the challenges of optical disaggregated data center networking for ML/AI applications

“…Due to the low latencies and high bit-rates that are required for hardware module to hardware module communications (up to several hundreds of Gb/s in the case of CPU-to-CPU or CPU-to-memory), high performance technologies are required. In this regard, optical network technologies are seen as prime candidates for the interconnection of the several hardware blades [6], [8]. Several optical switches as well as fiber links enable the interconnection within and between the multiple racks hosting the blades.…”

“…These impairments are especially notorious when dealing with the allocation of ML applications, since their accentuated time-varying behavior and their highly demanding resource profile pushes the limits of what traditional server units are capable of. In light of these limitations, the resource disaggregation paradigm has been proposed [6], [7]. It consists on separating the computational resources of server units in completely independent hardware blades, which then are hosted at rack scale.…”

A Novel Decomposed Optical Architecture for Satellite Terrestrial Network Edge Computing

“…A decomposed in-memory store framework was proposed in [18] for big data applications, utilizing theThymesisFlow memory system. A Software-Defined Networking (SDN)-based orchestration plane is designed and experimentally implemented for a decomposed computing architecture in [19] with hybrid optical switches. Decomposed architectures based on optical switches were proposed and experimentally investigated in [20,21], exploiting high aggregation bandwidths and transparent switching; however, the high switching delay (milliseconds) of the optical switch technology cannot provide nanoseconds scale communication between processor and memory cubes.…”

Integrating Resilient Tier N+1 Networks with Distributed Non-Recursive Cloud Model for Cyber-Physical Applications

“…Distributed cloud computing data center (DCCN) was applied in smart green Energy Internet [3]. The authors in [20] focused on a disaggregated application-centric optical network (DACON) for DC infrastructures using hybrid optical switches. Their work solved the current challenges regarding bottleneck performance issues and poor resource utilization in recent server centric DCNs.…”