Offloading load balancers onto SmartNICs

Cui, Tianyi; Zhang, Wei; Zhang, Kaiyuan

doi:10.1145/3476886.3477505

Cited by 13 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the evaluations, we observe that our implementation meets the processing rates required for control messages of offloaded applications. engines and runtime-configurable rules for routing packets [23].…”

Section: Discussionmentioning

confidence: 99%

“…In on-path smart-NICs, all traffic is handled by the NIC cores and tasks are executed on the smartNIC by adding logic to the processing pipeline [38]. The NIC cores are able to invoke special hardware accelerators for tasks such as crypto and compression [23]. In off-path smart-NICs, the NIC's cores are not directly on the data path from the host to the network, but instead, there exists a NIC-switch connecting the network ports, the host cores and the NIC cores [54].…”

Section: A Appendixmentioning

confidence: 99%

See 1 more Smart Citation

Implementing ChaCha based crypto primitives on programmable SmartNICs

Kottur

Kadiyala

Tammana

et al. 2022

Proceedings of the ACM SIGCOMM Workshop on Formal Foundations and Security of Programmable Network Infrastructures

View full text Add to dashboard Cite

Control and management plane applications such as serverless function orchestration and 4G/5G control plane functions are offloaded to smartNICs to reduce communication and processing latency. Such applications involve multiple inter-host interactions that were traditionally secured using SSL/TLS gRPC-based communication channels. Offloading the applications to smartNIC implies that we must also offload the security algorithms. Otherwise, we need to send the application messages to the host VM/container for crypto operations, negating offload benefits.We propose crypto externs for Netronome Agilio smartNICs that implement authentication and confidentiality (encryption/decryption) using the ChaCha stream cipher algorithm. AES and ChaCha are two popular cipher suites, but we chose ChaCha since none of the smartNICs have ChaCha-based crypto accelerators. However, smartNICs have restricted instruction set, and limited memory, making it difficult to implement security algorithms. This paper identifies and addresses several challenges to implement ChaCha crypto primitives successfully. Our evaluations show that our crypto extern implementation satisfies the scalability requirement of popular applications such as serverless management functions and host in-band network telemetry.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: A Appendixmentioning

confidence: 99%

Implementing ChaCha based crypto primitives on programmable SmartNICs

Kottur

Kadiyala

Tammana

et al. 2022

Proceedings of the ACM SIGCOMM Workshop on Formal Foundations and Security of Programmable Network Infrastructures

View full text Add to dashboard Cite

show abstract

“…Will Hardware Offloading Reduce Service Mesh Overheads? Another direction researchers are currently exploring is to offload the sidecar logic to a programmable network hardware [26]. While this is a promising direction, there are substantial challenges.…”

Section: Discussion and Future Workmentioning

confidence: 99%

“…Second, service mesh developers should be able to understand the impact of their optimizations on real-world applications. Today, there are several ongoing directions in industry and academia to reduce the service mesh performance overheads [5,26]. However, these service mesh developers do not have a way to get these insights (without doing their own extensive benchmarking experiments).…”

Section: Modeling Service Mesh Overheadmentioning

confidence: 99%

Dissecting Service Mesh Overheads

Zhu¹,

She²,

Xue³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Service meshes play a central role in the modern application ecosystem by providing an easy and flexible way to connect different services that form a distributed application. However, because of the way they interpose on application traffic, they can substantially increase application latency and resource consumption. We develop a decompositional approach and a tool, called MeshInsight, to systematically characterize the overhead of service meshes and to help developers quantify overhead in deployment scenarios of interest. Using MeshInsight, we confirm that service meshes can have high overhead-up to 185% higher latency and up to 92% more virtual CPU cores for our benchmark applications-but the severity is intimately tied to how they are configured and the application workload. The primary contributors to overhead vary based on the configuration too. IPC (inter-process communication) and socket writes dominate when the service mesh operates as a TCP proxy, but protocol parsing dominates when it operates as an HTTP proxy. MeshInsight also enables us to study the end-to-end impact of optimizations to service meshes. We show that not all seemingly-promising optimizations lead to a notable overhead reduction in realistic settings.

show abstract