LF_OVS_17_Red Hat's perspective on OVS HW Offload Status

•

1 like•1,741 views

This document summarizes Red Hat's perspective on the status of OVS hardware offloading. It discusses why offloading is needed to avoid using too many CPU cores for software switching. It provides examples of performance gains seen with various NIC vendors' offloading solutions integrated into the kernel and OVS. While many vendors now have offerings, more work remains to be done and is ongoing to fully integrate offloading capabilities.

Technology

Red Hat's perspective on
OVS HW Offload Status
Current state and what is WIP
Rashid Khan
Senior Manager, Networking Services
Nov. 17, 2017

INSERT DESIGNATOR, IF NEEDED
Acknowledgements and Disclaimers
I am presenting the work of many many people… Special thanks to:
Andrew T, Franck B, Eelco C, Marcelo L, Paolo A, Flavio L, Kevin T
Performance numbers shown in this presentation are based on test results from running
a specific series of tests in our labs.
Test results vary from one setup to another and based on different use cases.
Any test results mentioned are for example-only scenarios and are not conclusive nor a
recommendation of one vendor’s solution over another.

INSERT DESIGNATOR, IF NEEDED
AGENDA
Why offload ?
Does it look promising?
What is left to do?
Backup / more info
Please view Franck’s presentation from
Thursday 11:30am:
OVS-DPDK for NFV: go live feedback!
Please view Aaron Conole’s presentation
from Thursday 3:30pm:
Conntrack + OvS

INSERT DESIGNATOR, IF NEEDED
Why not just SW?
Simply way too many cores needed
●
○
●

Host
user land
kernel
vhost-user
OVS-DPDK
RX TX RX TX
user land
kernel
Guest: 5 vCPUs
virtio driver
RX TX
eth0
ssh, SNMP,
...
Virtio DPDK PMD
RX TXRX TX RX TXRX TX
CPU0 CPU1 CPU2 CPU3 CPU4
VNFc dataplane == stack
ACTIVE LOOP
while (1) {
RX-packet()
forward-packet()
}

de-queueing
en-queueing
vhostpci
polling thread
pcivhost
virtio virtio
10GSFP+Cu
●
●
●
●
●
●
●
●
9.38 6.30 4.69 3.15 9.15 6.15 4.57 3.07 7.39 4.97 3.69 2.48 4.34 2.92 2.17 1.46
7.66 16.19 3.83 8.45 7.81 3.90 8.66 7.45 6.52 3.72 3.26 2.34 1.17 2.59
2.39 19.99 1.19 9.99 2.39 1.19 9.99 2.39 19.99 1.19 9.99 2.38 1.19 9.96
1.64 19.99 0.82 9.99 1.64 1.64 19.99 0.82 9.89 1.64

INSERT DESIGNATOR, IF NEEDED
Very near future (some HW, some SW)

INSERT DESIGNATOR, IF NEEDED
NETRONOME MELLANOX
BROADCOM
Many HW vendors have OVS Offload solutions
CHELSIO
CAVIUM
Others

INSERT DESIGNATOR, IF NEEDED
Offloading method
Netronome TC (kernel)
Mellanox TC (kernel)
Broadcom TC (kernel)
Chelsio TC (kernel)
Cavium OVS runs in the NIC firmware,
offloading is transparent from CPU PoV
Accepted in upstream netdev

INSERT DESIGNATOR, IF NEEDED
Example
PVP test
SR-IOV
OVS
VM
Packet gen
(Xena / TRex)
TC layer
Kernel Driver
Virtual Function
Flow programming /
Slow path traffichttps://github.com/chaudron/ovs_perf
https://developers.redhat.com/blog/2017/0
9/28/automated-open-vswitch-pvp-testing/

INSERT DESIGNATOR, IF NEEDED
NETRONOME
Purple is
sw only

INSERT DESIGNATOR, IF NEEDED
What is WIP / To-do List
Rome wasn't built in a day…
●
●
●
●
●
●
●
●
●
●

THANK YOU
plus.google.com/+RedHat
linkedin.com/company/red-hat
youtube.com/user/RedHatVideos
facebook.com/redhatinc
twitter.com/RedHatNews
For further questions / comments:
rkhan@redhat.com
We Are Hiring !!!!

INSERT DESIGNATOR, IF NEEDED
SW used for testing
Netronome:
Linux upstream kernel, v4.13 for PVP test results. Linux V4.14rc4 for TC insertion rates.
OVS master branch from October 26th (7b997d4). DPDK/testpmd on VM v16.07
Mellanox:
Linux upstream kernel, net-next commit e1ea2f9856b7.
OVS master branch commit b05af21631ce, DPDK 17.11-rc2 (all git tips from Oct 30th).
OVS-DPDK:
RHEL7.4 latest kernel, OVS master branch from September 26th (97ee6d4), DPDK v17.05.2,
DPDK/testpmd on VM v16.07

This document discusses hardware offloading of VXLAN encapsulation and decapsulation in OVS-DPDK. It proposes representing virtual ports (vPorts) as tables to enable hardware offloading of VXLAN processing. Matching and actions on the vPort table would occur in hardware before decapsulation. Fallback processing using software would be used if full hardware offloading is not possible. The goal is to leverage intelligent NIC capabilities to accelerate VXLAN tunnel processing and improve performance for cloud, NFV, and storage workloads.

LF_OVS_17_Enabling hardware acceleration in OVS-DPDK using DPDK Framework.

LF_OpenvSwitch

The document discusses enabling hardware acceleration in OVS-DPDK using the DPDK 'Framework'. It describes challenges with hardware acceleration and introduces the DPDK Framework, which provides APIs and components to abstract different hardware features. The Framework is used in OVS-DPDK by initializing it for hardware offload, adding ports to the switch, reporting exception packets, installing flows in software and hardware pipelines. Next steps include publishing Framework APIs, getting early feedback, and implementing OVS-DPDK integration using the Framework.

LF_OVS_17_OVS Performance on Steroids - Hardware Acceleration Methodologies

LF_OpenvSwitch

This document discusses hardware acceleration methodologies for Open vSwitch (OVS) using Mellanox ConnectX network interface cards. It describes two approaches: full OVS offload using single-root I/O virtualization (SR-IOV) and partial offload using DPDK. Full offload moves the entire virtual switch to hardware, improving performance significantly over software-only approaches. Partial offload uses hardware for packet classification to accelerate parts of the OVS pipeline in DPDK. The document outlines performance benefits and future work areas like table offloading and live migration support. It also reviews community contributions to kernel, OVS, OpenStack and DPDK integration of these hardware offload techniques.

LF_OVS_17_Enabling Hardware Offload of OVS Control & Data plane using LiquidIO

LF_OpenvSwitch

This document discusses enabling Open vSwitch (OVS) hardware offload using Cavium's LiquidIO smart network interface cards (NICs). It describes two models for offloading OVS - data plane offload which keeps the control plane on the host, and full offload which moves both control and data planes to the NIC. The LiquidIO model represents a full offload where OVS runs natively on the NIC's processor. Performance tests show LiquidIO OVS offload achieving higher throughput and lower CPU usage than software-based OVS. Integration with OpenStack is also discussed.

LF_OVS_17_OvS manipulation with Go at DigitalOcean

LF_OpenvSwitch

The document discusses DigitalOcean's past and present use of Open vSwitch (OvS) for virtual networking. In the past, OvS was manipulated using Perl scripts that built flow strings and called ovs-ofctl. This had issues like lack of testing and non-atomic flow applications. Now, a Go package called ovs is used to programmatically control OvS. It builds flows without string manipulation and applies them atomically. DigitalOcean also uses packages like hvflow and gRPC services hvflowctl and hvflowd to configure OvS flows from network parameters. The future may involve orchestrating OvS directly through OpenFlow to avoid parsing tool outputs and directly applying flows.

LF_DPDK17_ OpenVswitch hardware offload over DPDK

LF_DPDK

This document discusses Mellanox Technologies' work on hardware offloading for Open vSwitch using DPDK. It describes concepts for using hardware classification offloading to accelerate Open vSwitch by offloading flow rules and classification to the NIC. It provides examples of performance improvements seen from using hardware classification offloading for Open vSwitch workloads. It also discusses approaches for using multi-table classification and hardware offloading for VxLAN encapsulation and decapsulation in Open vSwitch.

LF_OVS_17_Open vSwitch Offload: Conntrack and the Upstream Kernel

LF_OpenvSwitch

This document summarizes a presentation on conntrack offloading in Open vSwitch. It discusses the current approach of using Netfilter Conntrack in the OVS kernel, efforts to offload conntrack rules to a SmartNIC, and ongoing work to offload established conntrack flows without offloading the entire conntrack table. Key advantages of the latter approach include keeping initial flow decisions in the kernel for consistency while allowing unlimited offloaded flows handled by the SmartNIC.

OVS v OVS-DPDK

Md Safiyat Reza

This document discusses Open vSwitch (OVS) and how using Data Plane Development Kit (DPDK) can improve its performance. It notes that with standard OVS, there are many components between a virtual machine and physical networking that cause scalability and performance issues due to context switches. OVS-DPDK addresses this by using polling, hugepages, pinned CPUs, and userspace I/O to bypass the kernel and reduce overhead. The document shows that using DPDK can increase OVS throughput by over 8x and reduce latency by 30-37% compared to standard OVS.

This document discusses using LXC Linux containers over Open vSwitch (OVS) for container networking. It provides information on the Orabuntu-LXC project which builds and installs OVS and LXC to deploy containerized Oracle software. It describes how LXC 2.1.0 added explicit support for OVS and how OVS can be configured as a systemd service. It also discusses using OVS with containerized DNS/DHCP and sending container traffic over GRE tunnels.

LF_OVS_17_OVS-DPDK Installation and Gotchas

LF_OpenvSwitch

1) The document provides instructions for installing and configuring OVS DPDK on Ubuntu 17.04, including specifying hardware, installing prerequisites, configuring grub, identifying NIC ports, binding interfaces to DPDK drivers, setting up the OVS bridge and adding ports. 2) Key steps include reserving hugepages in grub, binding NICs to igb_uio or vfio-pci drivers, setting OVS configuration like datapath type and memory allocation, and adding interfaces to the OVS bridge. 3) The scripts provided automate many of these steps but additional manual configuration may still be needed and issues can occur with making interfaces persistent after reboots.

LF_OVS_17_Ingress Scheduling

LF_OpenvSwitch

OVS Hardware Offload with TC Flower

Netronome

DPDK Support for New HW Offloads

Netronome

This document discusses DPDK support for new hardware offloads. It describes the Netronome Agilio SmartNIC, which has hardware accelerators and can offload tasks like cryptography and flow processing. It discusses using the SmartNIC with DPDK and OVS for improved performance over kernel-based solutions. Full flow classification and action offloading to the SmartNIC is proposed to reduce CPU usage, along with exploring eBPF/XDP offloading possibilities and virtio offloading to enable VM migration.

Accelerate Service Function Chaining Vertical Solution with DPDK

OPNFV

Service Function Chaining (SFC) is one of top 5 NFV use case. Supporting SFC in provider and enterprise networks requires performance assurance. Specifically, the Classifier and the Service Function Forwarder which are typically implemented in software such as virtual switches need to match line rate requirement. DPDK (Data Plane Development Kit) is an open source project comprising a set of libraries and drivers for fast packet processing. In this presentation, we will discuss our experiences accelerating SFC with DPDK. In addition, Telco and Datacenter carriers demands dynamic SFC that requires new SFC wire protocols (e.g. VxLAN-GPE and NSH) support in both data and control planes. We intend to share our experiences and future works of a high performance, NSH-aware SFC vertical solution with open-source ingredients: Openstack, Opendaylight, OpenvSwitch with DPDK acceleration.

LF_OVS_17_State of the OVN

LF_OpenvSwitch

OVN has come a long way since its initial focus on providing virtual networking for OpenStack in a way that had minimal dependencies and complexity. Over several OpenStack releases, OVN has implemented key Neutron abstractions like logical switches, routers, security groups, and load balancing using OpenFlow. It provides distributed implementations of these functions along with new capabilities like ACL logging, DHCP services, and L3 gateway high availability. OVN is now reusable outside of OpenStack as well with integrations for Kubernetes, Docker, and other platforms.

Accelerating Neutron with Intel DPDK

Alexander Shalimov

Accelerating Neutron with Intel DPDK from #vBrownBag session at OpenStack Summit Atlanta 2014. 1. Many OpenStack deployments use Open vSwitch plugin for Neutron. 2. But its performance and scalability are not enough for production. 3. Intel DPDK vSwitch - an DPDK optimized version of Open vSwitch developed by Intel and publicly available at 01.org. But it doesn't have enough functionality for Neutron. We have implemented the needed parts included GRE and ARP stacks. Neutron pluging 4. We got 5 times performance improving for netwroking in OpenStack!

DPACC Acceleration Progress and Demonstration

OPNFV

The session provides an update to on the DPACC project within the OPNFV with a brief discussion on APIs and implementation progress. This session will review the API definition progress and follow up with a demo highlighting a common application as the vNF running on top of the DPACC defined layers. The demo will highlight the use of both hardware and software acceleration utilizing the DPACC defined acceleration layers. The demonstrationIt will highlight the progress in optimizing performance and latency characteristics of a platform to realize the vision of NFV while meeting stringent requirements, particularly for certain workloads, required by carriers.

LF_OVS_17_OVN and Containers - An update.

LF_OpenvSwitch

This document discusses integrating Open vSwitch (OVS) and Open Virtual Network (OVN) with containers and orchestrators like Docker and Kubernetes. It provides examples of using OVS and OVN commands to connect containers running in different namespaces to logical switches managed by OVN. It also describes how OVN implements Kubernetes networking concepts like pods, services, load balancing, and network policies using logical ports, switches and routers.

LF_DPDK17_Event Adapters - Connecting Devices to Eventdev

LF_DPDK

This document discusses event adapters in DPDK, which interface devices like network cards and cryptodevices to the event device framework. It describes Ethernet Rx, crypto, and timer adapters. The Ethernet Rx adapter handles receiving packets from network cards and enqueueing events. The crypto adapter supports enqueueing crypto operation events. The timer adapter provides a way to arm timers that generate expiry events. Event adapters provide common APIs and allow devices to produce events that are consumed by DPDK components running on multiple processor cores.

LF_DPDK17_DPDK support for new hardware offloads

LF_DPDK

This document discusses DPDK support for new hardware offloads. It describes the Netronome Agilio SmartNIC card, which has hardware accelerators and can offload tasks like cryptography and flow processing. It discusses using the card with OVS-DPDK for improved performance over kernel OVS. Full offloading of OVS classification and actions to the NIC is proposed to reduce CPU usage. Supporting eBPF offloading is also discussed, as is using DPDK with virtio. Overall the document explores how DPDK can better leverage network hardware offloads.

High Performance Networking Leveraging the DPDK and Growing Community

6WIND

ONIC Japan 2016 - Contrail アップデート

Juniper Networks (日本)

Using Network Acceleration for an Optimized Edge Cloud Server Architecture

Netronome

With the rise of cloud-native principles, applications are increasingly able to take advantage of diverse, specialized and distributed infrastructure. The emergence of Edge Cloud solutions promises faster and more immersive application experiences, as well as infrastructure primitives for 5G, IoT, mobility, and more. However, this new resource comes with space and power constraints that can only be overcome by using new disaggregated architectures that leverage network acceleration and optimally sized CPUs. The session will highlight how the capabilities unleashed by hardware offload of eBPF in edge cloud microservers will enable developers to efficiently leverage the massive amounts of data on the edge and to create next-generation real-time applications.

Open vSwitch Offload: Conntrack and the Upstream Kernel

Netronome

Offloading all or part of the Open vSwitch datapath to SmartNICs has been shown to not only release CPU resources on the server, but improve traffic processing performance. Recently steps have been made to support such offloading in the upstream Linux kernel. This has focused on creating an OVS datapath using the TC flower filter and utilizing the offload hooks already present here. This presentation focuses on how Connection Tracking (Conntrack) may fit into this model. It describes current work being undertaken with the Netfilter community to allow offloading of Conntrack entries. It continues to link this work with the offloading of Conntrack rules within OVS-TC.

DPDK Summit 2015 - Aspera - Charles Shiflett

Jim St. Leger

DPDK in Containers Hands-on Lab

Michelle Holley

This document provides an agenda and overview for a hands-on lab on using DPDK in containers. It introduces Linux containers and how they use fewer system resources than VMs. It discusses how containers still use the kernel network stack, which is not ideal for SDN/NFV usages, and how DPDK can be used in containers to address this. The hands-on lab section guides users through building DPDK and Open vSwitch, configuring them to work with containers, and running packet generation and forwarding using testpmd and pktgen Docker containers connected via Open vSwitch.

Network Jumbo Frame Config Guide

Woo Hyung Choi

LF_OVS_17_OvS Hardware Offload with TC Flower

LF_OpenvSwitch

LF_OVS_17_OVS-DPDK for NFV: go live feedback!

LF_OpenvSwitch

What's hot

DPDK Summit 2015 - HP - Al Sanders

Jim St. Leger

LF_OVS_17_LXC Linux Containers over Open vSwitch

LF_OpenvSwitch

LF_OVS_17_OVS-DPDK Installation and Gotchas

LF_OpenvSwitch

LF_OVS_17_Ingress Scheduling

LF_OpenvSwitch

OVS Hardware Offload with TC Flower

Netronome

DPDK Support for New HW Offloads

Netronome

Accelerate Service Function Chaining Vertical Solution with DPDK

OPNFV

LF_OVS_17_State of the OVN

LF_OpenvSwitch

Accelerating Neutron with Intel DPDK

Alexander Shalimov

DPACC Acceleration Progress and Demonstration

OPNFV

LF_OVS_17_OVN and Containers - An update.

LF_OpenvSwitch

LF_DPDK17_Event Adapters - Connecting Devices to Eventdev

LF_DPDK

LF_DPDK17_DPDK support for new hardware offloads

LF_DPDK

High Performance Networking Leveraging the DPDK and Growing Community

6WIND

ONIC Japan 2016 - Contrail アップデート

Juniper Networks (日本)

Using Network Acceleration for an Optimized Edge Cloud Server Architecture

Netronome

Open vSwitch Offload: Conntrack and the Upstream Kernel

Netronome

DPDK Summit 2015 - Aspera - Charles Shiflett

Jim St. Leger

DPDK in Containers Hands-on Lab

Michelle Holley

Network Jumbo Frame Config Guide

Woo Hyung Choi

What's hot (20)

DPDK Summit 2015 - HP - Al Sanders

LF_OVS_17_LXC Linux Containers over Open vSwitch

LF_OVS_17_OVS-DPDK Installation and Gotchas

LF_OVS_17_Ingress Scheduling

OVS Hardware Offload with TC Flower

DPDK Support for New HW Offloads

Accelerate Service Function Chaining Vertical Solution with DPDK

LF_OVS_17_State of the OVN

Accelerating Neutron with Intel DPDK

DPACC Acceleration Progress and Demonstration

LF_OVS_17_OVN and Containers - An update.

LF_DPDK17_Event Adapters - Connecting Devices to Eventdev

LF_DPDK17_DPDK support for new hardware offloads

High Performance Networking Leveraging the DPDK and Growing Community

ONIC Japan 2016 - Contrail アップデート

Using Network Acceleration for an Optimized Edge Cloud Server Architecture

Open vSwitch Offload: Conntrack and the Upstream Kernel

DPDK Summit 2015 - Aspera - Charles Shiflett

DPDK in Containers Hands-on Lab

Network Jumbo Frame Config Guide

Viewers also liked

LF_OVS_17_OvS Hardware Offload with TC Flower

LF_OpenvSwitch

LF_OVS_17_OVS-DPDK for NFV: go live feedback!

LF_OpenvSwitch

LF_OVS_17_OVN and Kelda

LF_OpenvSwitch

This document discusses the author's experiences using Open vSwitch Network Virtualization (OVN) with Kelda, a platform that encodes operational expertise in code. The author found OVN to be extremely stable but sees opportunities for improving ACL scaling and adding programming language support beyond C. Overall, OVN compares favorably to other networking solutions for containers but could benefit from more marketing efforts to increase awareness in the container community.

LF_OVS_17_Day 2 Closing Remarks

LF_OpenvSwitch

LF_OVS_17_The birth of SmartNICs -- offloading dataplane traffic to...software

LF_OpenvSwitch

LF_OVS_17_CORD: An open source platform to reinvent the network edge

LF_OpenvSwitch

CORD is an open platform for building virtualized network services at the edge of telecommunications networks. It uses open source software and white box hardware to provide a flexible edge cloud platform. Several major telecommunications providers worldwide have deployed or are planning to deploy CORD, including AT&T, Verizon, China Mobile and Deutsche Telekom. CORD comes in different variants like R-CORD for residential services, M-CORD for 5G mobile networks, and E-CORD for enterprise services. It aims to provide a programmable edge cloud platform with automated provisioning and operations.

LF_OVS_17_OVN at Nutanix

LF_OpenvSwitch

Nutanix uses a hyperconverged infrastructure to simplify hybrid cloud environments and manage workloads across private and public clouds. It offers a disaster recovery as a service solution called Xi Cloud to allow customers to recover workloads on public cloud if their on-premises infrastructure fails. Nutanix requires an SDN solution to provide overlay networking and features like microsegmentation across its hybrid environments. It is evaluating the OVN project for this purpose due to benefits like simpler abstractions, rich feature support, and integration with OVS, but also faces challenges with documentation, scaling limits, and feature parity. Nutanix plans to contribute to OVN by improving documentation, sharing reference architectures, and participating in code reviews and community efforts

LF_OVS_17_OVS/OVS-DPDK connection tracking for Mobile usecases

LF_OpenvSwitch

1) Mobile networks today handle a large number of simultaneous short duration flows, with high call rates of 100k-200k connections per second. Statistics like call duration and bandwidth usage need to be tracked for each flow for billing purposes. 2) Testing was conducted injecting a 10Gbps mobile traffic profile of 1 million flows into OVS, with 200k flows created and destroyed per second. Key metrics measured were maximum throughput, latency, and jitter at different flow table sizes and core counts. 3) Conntrack performance was tested for OVS kernel and DPDK versions. For 100k flows, OVS kernel achieved 152k pps for 4-tuple matching while OVS-DPDK achieved

LF_OVS_17_IPSEC and OVS DPDK

LF_OpenvSwitch

This document discusses proposed IPsec functionality for securing VXLAN traffic in a datacenter. It describes using IPsec in transport mode with AES-CBC and HMAC-SHA1-96 to provide confidentiality, integrity and authentication. A new "vxlanipsec" interface type is proposed to handle VXLAN encapsulation/decap and ESP encapsulation/decap using DPDK cryptodev for hardware acceleration. Performance metrics show encap rates of 2.7-7.1 million packets per second for a single PMD instance on Intel hardware. Future work includes supporting GCM mode, IPsec tunnels, dynamic key re-keying and integrating with OVS and RTE_Security.

LF_OVS_17_Day 2 Opening Remarks

LF_OpenvSwitch

The document provides the agenda for Day 2 of the Open vSwitch 2017 Fall Conference. It outlines that the day will include a keynote in the morning followed by multiple sessions on various technical topics related to Open vSwitch. It also provides instructions for speakers, including submitting slides to the AV desk in advance and paying attention to timers during talks. Finally, it notes that extra conference t-shirts are available.

LF_OVS_17_DigitalOcean Cloud Firewalls: powered by OvS and conntrack

LF_OpenvSwitch

Viewers also liked (11)

LF_OVS_17_OvS Hardware Offload with TC Flower

LF_OVS_17_OVS-DPDK for NFV: go live feedback!

LF_OVS_17_OVN and Kelda

LF_OVS_17_Day 2 Closing Remarks

LF_OVS_17_The birth of SmartNICs -- offloading dataplane traffic to...software

LF_OVS_17_CORD: An open source platform to reinvent the network edge

LF_OVS_17_OVN at Nutanix

LF_OVS_17_OVS/OVS-DPDK connection tracking for Mobile usecases

LF_OVS_17_IPSEC and OVS DPDK

LF_OVS_17_Day 2 Opening Remarks

LF_OVS_17_DigitalOcean Cloud Firewalls: powered by OvS and conntrack

Similar to LF_OVS_17_Red Hat's perspective on OVS HW Offload Status

Why Use Oracle VM for Oracle Databases? Revera Presentation

Francisco Alvarez

Hardware accelerated switching with Linux @ SWLUG Talks May 2014

Nat Morris

Introduzione a Software Define Networking

festival ICT 2016

This document provides an introduction to software-defined networking (SDN) concepts. It defines SDN as separating the control plane and data plane in network devices to make network implementation, scalability, and management easier. The document discusses SDN architectures using OpenFlow to define communication between the control and forwarding layers. It also covers SDN implementations using overlays like VXLAN and challenges around SDN protocols, scalability, and performance.

Dev ops with smell v1.2

Antons Kranga

Introduction to nfv movilforum

videos

This document provides an introduction to network functions virtualization (NFV) and discusses its potential benefits and challenges. Some key points: - NFV involves separating network functions from proprietary hardware appliances and implementing them as software virtual network functions (VNFs) that can run on standard server hardware. - This allows network functions to be deployed flexibly on commodity hardware and moved easily between data centers. It also aims to reduce costs and simplify network operations. - Integration with legacy systems and ensuring interoperability between VNFs are seen as main challenges. Data plane performance is also critical for the most demanding use cases. - Software defined networking (SDN) helps control and interconnect VNFs by

Corwin on Containers

Corwin Brown

This was a talk given internally at BloomReach as well as a guest lecture to a grad level Data Structures and Algorithms class at the University of Texas at Arlington. A Deep dive on the history of containers, and how they work under the cover utilizing Linux Kernel features such as Process Namespaces and Control Groups. I also go over a bit of the history of Container technology, going from Chroot and Jails and Zones, to LXC and Docker

MesosCon_Asia_2016_DevOps_at_GSShop

Vivek Juneja

This document discusses the transition from traditional operations and development practices to a DevOps model over time. It begins with the challenges of the past state including long lead times, lack of collaboration between teams, and reliance on big bang releases. The beginning of the change involved adopting principles like infrastructure as code and shared tooling/platforms to improve productivity. Fully adopting DevOps required establishing new roles, processes, and addressing cultural issues. The road ahead focuses on continuous improvement areas like architecture constraints, multi-tenant clusters, and advanced application deployment techniques. The overall summary discusses the evolution from initial changes to mature DevOps practices.

Summit 16: How to Compose a New OPNFV Solution Stack?

OPNFV

This session showcases how a new OPNFV solution stack (a.k.a. ""scenario"") is composed and stood up. We'll use a new solution stack framed around a new software forwarder (""VPP"") provided by the FD.io project as example for this session. The session discusses how an evolution/change of upstream components from OpenStack, OpenDaylight and FFD.io are put in place for the scenario, how installers and tests need to be evolved to allow for integration into OPNFV's continuous integration, deployment and test pipeline.

[KubeCon NA 2018] Telepresence Deep Dive Session - Rafael Schloming & Luke Sh...

Ambassador Labs

One of the challenges facing Telepresence is growing the contributor community. It’s a complex application that requires a good understanding of OS networking, VPNs, Kubernetes, and everything in between. We’ll kick off this meeting with a general architectural overview of Telepresence. We’ll talk about how we’ve managed the project to date, and our investments to make it easier. We want to then turn it over for an interactive discussion with participants to see what we can do to make it easier to contribute and grow the Telepresence community.

DoS and DDoS mitigations with eBPF, XDP and DPDK

Marian Marinov

The document compares eBPF, XDP and DPDK for packet inspection. It describes the speaker's experience using these tools to build a virtual machine that can handle 10Gbps of traffic and drop packets to mitigate DDoS attacks. It details how eBPF and XDP were able to achieve higher packet drop rates than iptables or a custom module. While DPDK could drop traffic at line rate, it required specialized hardware and expertise. Ultimately, XDP provided the best balance of performance, driver support and programmability using eBPF to drop millions of packets per second.

200-301-demo.pdf

CiscoExamDumpsarticl1

Cisco 200-301 Exam Dumps

CiscoExamDumpsarticl2

How is this newsletter going to help you? Apart from providing you with a brief glimpse of the test’s topics and shape, we are able to additionally assist you discover efficient education substances. Cisco’s internet site is a extraordinary starting point, but you shouldn’t restriction your self to it. Despite the fact that you would possibly have by no means heard approximately them, you have to attempt exam dumps as they may grow to be your secret tool to get a passing rating in 2 hundred-301 assessment. But now, let’s start with the exam details.

Cisco 200-301 Exam Dumps

CiscoExamDumpsarticl

OSDC 2014 ONIE by Nat Morris

Cumulus Networks

ONIE (Open Network Install Environment) provides an environment for installing network operating systems on bare-metal switches. It is implemented using a Linux kernel and BusyBox. ONIE configures the management interfaces and locates and executes OS installers from the network or USB. It allows choice of hardware vendors, operating system vendors, and provides multi-vendor interoperability. ONIE is an open source project within the Open Compute Project.

Optimizing, Profiling, and Deploying TensorFlow AI Models with GPUs - San Fra...

Chris Fregly

http://pipeline.ai Using the latest advancements from TensorFlow including the Accelerated Linear Algebra (XLA) Framework, JIT/AOT Compiler, and Graph Transform Tool, I’ll demonstrate how to optimize, profile, and deploy TensorFlow Models - and the TensorFlow Runtime - in GPU-based production environment. This talk is 100% demo based on open source tools and completely reproducible through Docker on your own GPU cluster. Bio Chris Fregly is Founder and Research Engineer at PipelineAI, a Streaming Machine Learning and Artificial Intelligence Startup based in San Francisco. He is also an Apache Spark Contributor, a Netflix Open Source Committer, founder of the Global Advanced Spark and TensorFlow Meetup, author of the O’Reilly Training and Video Series titled, "High-Performance TensorFlow in Production." Previously, Chris was a Distributed Systems Engineer at Netflix, a Data Solutions Engineer at Databricks, and a Founding Member and Principal Engineer at the IBM Spark Technology Center in San Francisco. http://pipeline.ai

OpenNebulaConf 2016 - OpenNebula, a story about flexibility and technological...

OpenNebula Project

Cloud providers are constantly addressing the technology limitations on their infrastructures, which must be overcome to meet customer needs. On this presentation, we will demonstrate how technological agnosticism and management flexibility of OpenNebula has allowed Todoencloud to provide the most efficient open source solution to the needs of its customers, choosing the most appropriate virtualization technology (Xen and KVM), storage approach (ZFS vs CEPH), Cloud Bursting solutions (Azure, Amazon) and customized networking topologies.

44CON 2014 - Stupid PCIe Tricks, Joe Fitzpatrick

44CON

Joe FitzPatrick gave a presentation on exploiting PCIe (Peripheral Component Interconnect Express) buses for hardware attacks. He discussed using DMA (direct memory access) over PCIe to read and write system memory, modify firmware, and potentially bypass mitigations like IOMMU (input-output memory management unit). FitzPatrick demonstrated proof-of-concept attacks on Macs and Windows PCs using custom PCIe devices and software. However, he noted that fully bypassing protections like VT-d on Macbooks had not yet been achieved and more work is needed to build attacks without imitating a genuine device.

Whitebox Switches Deployment Experience

APNIC

ZCX Hybrid Application

Michael Erichsen

Development, test, and characterization of MEC platforms with Teranium and Dr...

Michelle Holley

Mobile edge computing delivers cloud computing at the edge of the cellular network to drive services quality and innovation. The ability for CSPs and ISVs to effectively develop, deliver, and deploy MEC services on a given platform directly correlates with the availability and maturity of associated tools and test environment. Dronava is a hyper-connected, web-scale network reference design for the 5G mobile network, suitable for use as a test and development socket for cloud applications developed for MEC platforms with tools such as the Intel NEV SDK. With Dronava, developers can drive the application with real traffics from the network edge to the EPC core, and if need be, connect with services in the core network in order to fully characterize the functionalities, latency, and throughput of the platform and application.Teranium is an integrated development environment that simplifies the development, packaging, and deployment/management of cloud applications. Teranium can be utilized to develop and deploy MEC applications on a number of platforms. Together with Dronava, Teranium helps to reduce complexity and improve efficiency in the ability of CSPs and ISVs to adopt and deploy MEC-base services.

Similar to LF_OVS_17_Red Hat's perspective on OVS HW Offload Status (20)

Why Use Oracle VM for Oracle Databases? Revera Presentation

Hardware accelerated switching with Linux @ SWLUG Talks May 2014

Introduzione a Software Define Networking

Dev ops with smell v1.2

Introduction to nfv movilforum

Corwin on Containers

MesosCon_Asia_2016_DevOps_at_GSShop

Summit 16: How to Compose a New OPNFV Solution Stack?

[KubeCon NA 2018] Telepresence Deep Dive Session - Rafael Schloming & Luke Sh...

DoS and DDoS mitigations with eBPF, XDP and DPDK

200-301-demo.pdf

Cisco 200-301 Exam Dumps

OSDC 2014 ONIE by Nat Morris

Optimizing, Profiling, and Deploying TensorFlow AI Models with GPUs - San Fra...

OpenNebulaConf 2016 - OpenNebula, a story about flexibility and technological...

44CON 2014 - Stupid PCIe Tricks, Joe Fitzpatrick

Whitebox Switches Deployment Experience

ZCX Hybrid Application

Development, test, and characterization of MEC platforms with Teranium and Dr...

Recently uploaded

UiPath Test Automation using UiPath Test Suite series, part 5

DianaGray10

Mind map of terminologies used in context of Generative AI

Kumud Singh

20240607 QFM018 Elixir Reading List May 2024

Matthew Sinclair

Driving Business Innovation: Latest Generative AI Advancements & Success Story

Safe Software

Are you ready to revolutionize how you handle data? Join us for a webinar where we’ll bring you up to speed with the latest advancements in Generative AI technology and discover how leveraging FME with tools from giants like Google Gemini, Amazon, and Microsoft OpenAI can supercharge your workflow efficiency. During the hour, we’ll take you through: Guest Speaker Segment with Hannah Barrington: Dive into the world of dynamic real estate marketing with Hannah, the Marketing Manager at Workspace Group. Hear firsthand how their team generates engaging descriptions for thousands of office units by integrating diverse data sources—from PDF floorplans to web pages—using FME transformers, like OpenAIVisionConnector and AnthropicVisionConnector. This use case will show you how GenAI can streamline content creation for marketing across the board. Ollama Use Case: Learn how Scenario Specialist Dmitri Bagh has utilized Ollama within FME to input data, create custom models, and enhance security protocols. This segment will include demos to illustrate the full capabilities of FME in AI-driven processes. Custom AI Models: Discover how to leverage FME to build personalized AI models using your data. Whether it’s populating a model with local data for added security or integrating public AI tools, find out how FME facilitates a versatile and secure approach to AI. We’ll wrap up with a live Q&A session where you can engage with our experts on your specific use cases, and learn more about optimizing your data workflows with AI. This webinar is ideal for professionals seeking to harness the power of AI within their data management systems while ensuring high levels of customization and security. Whether you're a novice or an expert, gain actionable insights and strategies to elevate your data processes. Join us to see how FME and AI can revolutionize how you work with data!

Infrastructure Challenges in Scaling RAG with Custom AI models

Zilliz

Building Retrieval-Augmented Generation (RAG) systems with open-source and custom AI models is a complex task. This talk explores the challenges in productionizing RAG systems, including retrieval performance, response synthesis, and evaluation. We’ll discuss how to leverage open-source models like text embeddings, language models, and custom fine-tuned models to enhance RAG performance. Additionally, we’ll cover how BentoML can help orchestrate and scale these AI components efficiently, ensuring seamless deployment and management of RAG systems in the cloud.

HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU

panagenda

Webinar Recording: https://www.panagenda.com/webinars/hcl-notes-und-domino-lizenzkostenreduzierung-in-der-welt-von-dlau/ DLAU und die Lizenzen nach dem CCB- und CCX-Modell sind für viele in der HCL-Community seit letztem Jahr ein heißes Thema. Als Notes- oder Domino-Kunde haben Sie vielleicht mit unerwartet hohen Benutzerzahlen und Lizenzgebühren zu kämpfen. Sie fragen sich vielleicht, wie diese neue Art der Lizenzierung funktioniert und welchen Nutzen sie Ihnen bringt. Vor allem wollen Sie sicherlich Ihr Budget einhalten und Kosten sparen, wo immer möglich. Das verstehen wir und wir möchten Ihnen dabei helfen! Wir erklären Ihnen, wie Sie häufige Konfigurationsprobleme lösen können, die dazu führen können, dass mehr Benutzer gezählt werden als nötig, und wie Sie überflüssige oder ungenutzte Konten identifizieren und entfernen können, um Geld zu sparen. Es gibt auch einige Ansätze, die zu unnötigen Ausgaben führen können, z. B. wenn ein Personendokument anstelle eines Mail-Ins für geteilte Mailboxen verwendet wird. Wir zeigen Ihnen solche Fälle und deren Lösungen. Und natürlich erklären wir Ihnen das neue Lizenzmodell. Nehmen Sie an diesem Webinar teil, bei dem HCL-Ambassador Marc Thomas und Gastredner Franz Walder Ihnen diese neue Welt näherbringen. Es vermittelt Ihnen die Tools und das Know-how, um den Überblick zu bewahren. Sie werden in der Lage sein, Ihre Kosten durch eine optimierte Domino-Konfiguration zu reduzieren und auch in Zukunft gering zu halten. Diese Themen werden behandelt - Reduzierung der Lizenzkosten durch Auffinden und Beheben von Fehlkonfigurationen und überflüssigen Konten - Wie funktionieren CCB- und CCX-Lizenzen wirklich? - Verstehen des DLAU-Tools und wie man es am besten nutzt - Tipps für häufige Problembereiche, wie z. B. Team-Postfächer, Funktions-/Testbenutzer usw. - Praxisbeispiele und Best Practices zum sofortigen Umsetzen

TrustArc Webinar - 2024 Global Privacy Survey

TrustArc

How does your privacy program stack up against your peers? What challenges are privacy teams tackling and prioritizing in 2024? In the fifth annual Global Privacy Benchmarks Survey, we asked over 1,800 global privacy professionals and business executives to share their perspectives on the current state of privacy inside and outside of their organizations. This year’s report focused on emerging areas of importance for privacy and compliance professionals, including considerations and implications of Artificial Intelligence (AI) technologies, building brand trust, and different approaches for achieving higher privacy competence scores. See how organizational priorities and strategic approaches to data security and privacy are evolving around the globe. This webinar will review: - The top 10 privacy insights from the fifth annual Global Privacy Benchmarks Survey - The top challenges for privacy leaders, practitioners, and organizations in 2024 - Key themes to consider in developing and maintaining your privacy program

Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf

Malak Abu Hammad

Discover how MongoDB Atlas and vector search technology can revolutionize your application's search capabilities. This comprehensive presentation covers: * What is Vector Search? * Importance and benefits of vector search * Practical use cases across various industries * Step-by-step implementation guide * Live demos with code snippets * Enhancing LLM capabilities with vector search * Best practices and optimization strategies Perfect for developers, AI enthusiasts, and tech leaders. Learn how to leverage MongoDB Atlas to deliver highly relevant, context-aware search results, transforming your data retrieval process. Stay ahead in tech innovation and maximize the potential of your applications. #MongoDB #VectorSearch #AI #SemanticSearch #TechInnovation #DataScience #LLM #MachineLearning #SearchTechnology

Mariano G Tinti - Decoding SpaceX

Mariano Tinti

Serial Arm Control in Real Time Presentation

tolgahangng

Uni Systems Copilot event_05062024_C.Vlachos.pdf

Uni Systems S.M.S.A.

GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024

Neo4j

Pushing the limits of ePRTC: 100ns holdover for 100 days

Adtran

Introduction to CHERI technology - Cybersecurity

mikeeftimakis1

20240609 QFM020 Irresponsible AI Reading List May 2024

Matthew Sinclair

Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack

shyamraj55

Communications Mining Series - Zero to Hero - Session 1

DianaGray10

This session provides introduction to UiPath Communication Mining, importance and platform overview. You will acquire a good understand of the phases in Communication Mining as we go over the platform with you. Topics covered: • Communication Mining Overview • Why is it important? • How can it help today’s business and the benefits • Phases in Communication Mining • Demo on Platform overview • Q/A

GenAI Pilot Implementation in the organizations

kumardaparthi1024

GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...

Neo4j

Sudheer Mechineni, Head of Application Frameworks, Standard Chartered Bank Discover how Standard Chartered Bank harnessed the power of Neo4j to transform complex data access challenges into a dynamic, scalable graph database solution. This keynote will cover their journey from initial adoption to deploying a fully automated, enterprise-grade causal cluster, highlighting key strategies for modelling organisational changes and ensuring robust disaster recovery. Learn how these innovations have not only enhanced Standard Chartered Bank’s data infrastructure but also positioned them as pioneers in the banking sector’s adoption of graph technology.

GraphRAG for Life Science to increase LLM accuracy

Tomaz Bratanic

Recently uploaded (20)

UiPath Test Automation using UiPath Test Suite series, part 5

Mind map of terminologies used in context of Generative AI

20240607 QFM018 Elixir Reading List May 2024

Driving Business Innovation: Latest Generative AI Advancements & Success Story

Infrastructure Challenges in Scaling RAG with Custom AI models

HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU

TrustArc Webinar - 2024 Global Privacy Survey

Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf

Mariano G Tinti - Decoding SpaceX

Serial Arm Control in Real Time Presentation

Uni Systems Copilot event_05062024_C.Vlachos.pdf

GraphSummit Singapore | The Art of the Possible with Graph - Q2 2024

Pushing the limits of ePRTC: 100ns holdover for 100 days

Introduction to CHERI technology - Cybersecurity

20240609 QFM020 Irresponsible AI Reading List May 2024

Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack

Communications Mining Series - Zero to Hero - Session 1

GenAI Pilot Implementation in the organizations

GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...

GraphRAG for Life Science to increase LLM accuracy

LF_OVS_17_Red Hat's perspective on OVS HW Offload Status

1. Red Hat's perspective on OVS HW Offload Status Current state and what is WIP Rashid Khan Senior Manager, Networking Services Nov. 17, 2017

2. INSERT DESIGNATOR, IF NEEDED Acknowledgements and Disclaimers I am presenting the work of many many people… Special thanks to: Andrew T, Franck B, Eelco C, Marcelo L, Paolo A, Flavio L, Kevin T Performance numbers shown in this presentation are based on test results from running a specific series of tests in our labs. Test results vary from one setup to another and based on different use cases. Any test results mentioned are for example-only scenarios and are not conclusive nor a recommendation of one vendor’s solution over another.

3. INSERT DESIGNATOR, IF NEEDED AGENDA Why offload ? Does it look promising? What is left to do? Backup / more info Please view Franck’s presentation from Thursday 11:30am: OVS-DPDK for NFV: go live feedback! Please view Aaron Conole’s presentation from Thursday 3:30pm: Conntrack + OvS

4. INSERT DESIGNATOR, IF NEEDED Why not just SW? Simply way too many cores needed ● ○ ●

5. Host user land kernel vhost-user OVS-DPDK RX TX RX TX user land kernel Guest: 5 vCPUs virtio driver RX TX eth0 ssh, SNMP, ... Virtio DPDK PMD RX TXRX TX RX TXRX TX CPU0 CPU1 CPU2 CPU3 CPU4 VNFc dataplane == stack ACTIVE LOOP while (1) { RX-packet() forward-packet() }

6. de-queueing en-queueing vhostpci polling thread pcivhost virtio virtio 10GSFP+Cu ● ● ● ● ● ● ● ● 9.38 6.30 4.69 3.15 9.15 6.15 4.57 3.07 7.39 4.97 3.69 2.48 4.34 2.92 2.17 1.46 7.66 16.19 3.83 8.45 7.81 3.90 8.66 7.45 6.52 3.72 3.26 2.34 1.17 2.59 2.39 19.99 1.19 9.99 2.39 1.19 9.99 2.39 19.99 1.19 9.99 2.38 1.19 9.96 1.64 19.99 0.82 9.99 1.64 1.64 19.99 0.82 9.89 1.64

7. INSERT DESIGNATOR, IF NEEDED

8. INSERT DESIGNATOR, IF NEEDED Very near future (some HW, some SW)

9. INSERT DESIGNATOR, IF NEEDED NETRONOME MELLANOX BROADCOM Many HW vendors have OVS Offload solutions CHELSIO CAVIUM Others

10. INSERT DESIGNATOR, IF NEEDED Offloading method Netronome TC (kernel) Mellanox TC (kernel) Broadcom TC (kernel) Chelsio TC (kernel) Cavium OVS runs in the NIC firmware, offloading is transparent from CPU PoV Accepted in upstream netdev

11. INSERT DESIGNATOR, IF NEEDED Example PVP test SR-IOV OVS VM Packet gen (Xena / TRex) TC layer Kernel Driver Virtual Function Flow programming / Slow path traffichttps://github.com/chaudron/ovs_perf https://developers.redhat.com/blog/2017/0 9/28/automated-open-vswitch-pvp-testing/

12. INSERT DESIGNATOR, IF NEEDED NETRONOME

13. INSERT DESIGNATOR, IF NEEDED Mellanox

14. INSERT DESIGNATOR, IF NEEDED NETRONOME Purple is sw only

15. INSERT DESIGNATOR, IF NEEDED Mellanox

16. INSERT DESIGNATOR, IF NEEDED What is WIP / To-do List Rome wasn't built in a day… ● ● ● ● ● ● ● ● ● ●

17. THANK YOU plus.google.com/+RedHat linkedin.com/company/red-hat youtube.com/user/RedHatVideos facebook.com/redhatinc twitter.com/RedHatNews For further questions / comments: rkhan@redhat.com We Are Hiring !!!!

18. More information

19. INSERT DESIGNATOR, IF NEEDED SW used for testing Netronome: Linux upstream kernel, v4.13 for PVP test results. Linux V4.14rc4 for TC insertion rates. OVS master branch from October 26th (7b997d4). DPDK/testpmd on VM v16.07 Mellanox: Linux upstream kernel, net-next commit e1ea2f9856b7. OVS master branch commit b05af21631ce, DPDK 17.11-rc2 (all git tips from Oct 30th). OVS-DPDK: RHEL7.4 latest kernel, OVS master branch from September 26th (97ee6d4), DPDK v17.05.2, DPDK/testpmd on VM v16.07

LF_OVS_17_Red Hat's perspective on OVS HW Offload Status

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (11)

Similar to LF_OVS_17_Red Hat's perspective on OVS HW Offload Status

Similar to LF_OVS_17_Red Hat's perspective on OVS HW Offload Status (20)

Recently uploaded

Recently uploaded (20)

LF_OVS_17_Red Hat's perspective on OVS HW Offload Status