Unleashing GPUs for Network Function Virtualization: an open architecture based on Vulkan and Kubernetes

  title={Unleashing GPUs for Network Function Virtualization: an open architecture based on Vulkan and Kubernetes},
  author={Juuso Haavisto and Thibault Cholez and Jukka Riekki},
  journal={NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium},
General-purpose computing on graphics processing units (GPGPU) is a promising way to speed up computationally intensive network functions, such as performing real-time traffic classification based on machine learning. Recent studies have focused on integrated graphics units and various performance optimizations to address bottlenecks such as latency. However, these approaches tend to produce architecture-specific binaries and lack the orchestration of functions. A complementary effort would be… 

Figures from this paper



Interoperable GPU Kernels as Latency Improver for MEC

This work proposes a client-server implementation for transacting intermediate representation (IR) between a mobile UE and a MEC server instead of video codecs and finds that due to low cold-start times on both UEs and MEC servers, application migration can happen in milliseconds.

Legate NumPy: accelerated and distributed array computing

Legate is introduced, a drop-in replacement for NumPy that requires only a single-line code change and can scale up to an arbitrary number of GPU accelerated nodes and achieve speed-ups of up to 10X on 1280 CPUs and 100X on 256 GPUs.

Transparent and Service-Agnostic Monitoring of Encrypted Web Traffic

The solution, H2Classifier, aims at detecting if a user performs an action that has been previously defined over a monitored Web service, but without using any decryption, based on passive traffic analysis and relies on random forest classifier.

GoldenEye: stream-based network packet inspection using GPUs

GoldenEye is presented, a deep packet inspection (DPI) system that tracks out-of-order TCP packets and provides stream-based signature matching and results show that GoldenEye can reassemble tens of millions of packets/sec and conduct stateful DPI operations on TCP streams at multi-ten Gbit/sec rates.

GEN: A GPU-Accelerated Elastic Framework for NFV

This work proposes GEN, a GPU-based high performance and elastic framework for NFV, which proposes to support RTC-based SFCs to improve processing performance and offers great elasticity of network function (NF) scaling up and down by allocating a different number of fine-grained GPU threads to an NF during runtime.

An Efficient GPU-Based Multiple Pattern Matching Algorithm for Packet Filtering

This paper proposes a GPU-based multiple-pattern matching algorithm for filtering malicious packets by using a Bloom filter to inspect the packet payload by leveraging the high parallelism computing power of GPU.

GPU Triggered Networking for Intra-KerneI Communications

This paper proposes GPU Triggered Networking, a novel, GPU-centric networking approach which leverages the best of CPUs and GPUs and illustrates how this approach can provide up to 25% speedup compared to standard GPU networking across microbenchmarks, a Jacobi stencil, an important MPI collective operation, and machine-learning workloads.

ASAP: As Static As Possible memory management

This dissertation develops asap: a new memory management technique that fits in the static-automatic gap, and establishes a precise and powerful lexicon to describe memory management strategies of any kind.

GPUNFV: a GPU-Accelerated NFV System

GPUNFV is presented, a high-performance NFV system providing flow-level micro services for stateful service chains with Graphics Processing Unit (GPU) acceleration that achieves a much better throughput than the existing NFV systems.

APUNet: Revitalizing GPU as Packet Processing Accelerator

The claim that CPU can outperform or achieve a similar performance as GPU if its code is re-arranged to run concurrently with memory access, employing optimization techniques such as group prefetching and software pipelining is revisited and seen if it can be generalized to a large class of network applications.