SlideShare a Scribd company logo
1 of 27
Download to read offline
Making Workload Nomadic When
Accelerated
Nomad Project Introduction
Zhipeng Huang - Huawei
Michele Paolino - Virtual Open Systems
Hui Deng - China Mobile
Outline
❏ Motivation
❏ Interesting Use Cases
❏ OPNFV Requirements
❏ Nomad Introduction
❏ Future Plan
Motivation
Motivation - Why ?
OpenStack currently lacks of a set of features in respect to portable hardware
acceleration*:
● Accelerators life management
● Accelerators resource discovery
● Reconfigurable FPGA, GPUs and other accelerators migration support, ease
of use, etc.
* these features have been highlighted in the OPNFV OpenStack GAP Analysis document.
The main purpose of Nomad is to implement all these
features, providing (all) the OpenStack users with a
portable, powerful, easy way to use hardware acceleration
Motivation - What ?
3 models if related to a more traditional CPU as part of the workload
Very Close - in the cpu chipset or on the board (i.e., see Intel’s skylake staging)
● Suitable for offload model, and inline if associated interface is also in place
● With optimal sharing of resources can provide excellent processing gains
● Limited by horizontal scale, but can be leveraged as a unit of management like the associated CPU
Nearby - attached via a bus or similar (i.e., PCIe or within a chassis assembly)
● Suitable for offload and inline models
● Susceptible to negative impact if interface across the bus is chatty
● Larger scale possible, particularly in chassis configurations
Far - reachable by TCP/IP or other communication protocol
● Suitable for offload and inline if latency is not a concern
● Largest horizontal scale flexibility
● Much more suited to a standalone function model
Motivation - How ?
Nova Extension Solely Nomad (Dedicated Management Function)
Target Best performance, no portability. Best performance/portability trade off.
Accelerator
access
Direct management. Management through a portability layer.
Pros - Direct interaction between the compute
node and the accelerators could provide
slightly better performance
- Resource discovery, scheduling, setup etc.
- Support for automatic integrated acceleration
management for accelerated VM's Migration
- Hardware portability and independence.
Cons - Portability/migration hard to support
- Code complexity: specific code needed
for each accelerator type, with impact to
the project performance, security,
maintainability, etc.
- The accelerator allocation phase might take
time, as an handshake procedure has to be put
in place
- Scalability issues
Outline
✓ Motivation
❏ Interesting Use Cases
❏ OPNFV Requirements
❏ Nomad Introduction
❏ Future Plan
Interesting Use Cases - NFVIaaS
● For people that are familiar with ETSI NFV standard,
NFVIaaS was among the NFV use cases that
published in Phase 1 documents
● However few of us grasp what this use case actually
meant for business, until now.
● Now many operators begin to build their own Public
Cloud. If an operator has a multiple site supported
Public Cloud, it then would be able to offer NFVIaaS to
NFV app companies that has its own content, or VNF,
or MANO, but without NFVI. These NFV apps then
could be deployed on the Public Cloud.
● There is still a problem for the operator who owns
the Public Cloud - so further service classification on
NFVIaaS without acceleration
Acceleration makes NFVIaaS a possibility
Interesting Use Cases - accelerated vSwitch
The next generation of virtual switch technologies will
combine data plane and low level SW acceleration with
network engines, PCI accelerators, etc:
● Dynamically allocating (reconfigurable)
accelerators and supporting VM migration
● Dispatching packet processing workloads on the
available resources
*http://www.virtualopensystems.com/en/products/vosyswitch-nfv-virtual-switch/
VOSYSwitch* is a user space virtual switch for NFV
(and not only), based on the framework Snabb
(snabb.io), and thus independent from Intel/DPDK:
● It inherits from Snabb the LuaJIT trace
compiler, which provides performance
optimization at runtime based on the current
traffic
● Supports x86/ARM architectures and
hardware accelerators through ODP
● Its roadmap includes FPGA/GPU/OpenCL
support, and integration with OpenStack
Nomad
Outline
✓ Motivation
✓ Interesting Use Cases
❏ OPNFV Requirements
❏ Nomad Introduction
❏ Future Plan
OPNFV Requirements
DPACC general
requirments : https://docs.
google.
com/document/d/1YexfnL
RZ9gZnj-
5PNOnrJ5CVMhbzZ7mIJo
gW6nyIGq0/edit
DPACC OpenStack
requirements : https:
//docs.google.
com/document/d/1_fOinIQ
NcPwNODZPzGK5vRMPJ
QLwL7iLds4NFnjXSms/ed
it
Outline
✓ Motivation
✓ Interesting Use Cases
✓ OPNFV Requirements
❏ Nomad Introduction
❏ Future Plan
Nomad Introduction - some really high level stuff
Design Goal:
● Loosely coupled (more than
Neutron)
● DB oriented management
● Network: DAG based
● Storage: affinity driven
● Compute : capability based
scheduling
● Big Data: FPGA/GPU programming
● Fine grained fault monitoring and
management (bloom filter?)
Nomad introduction - first stab at nomad-compute
● BP Driven, everything
starts from scratch
● Target code available
in August
● Try to make OPNFV
Colorado release if
possible
Outline
✓ Motivation
✓ Interesting Use Cases
✓ OPNFV Requirements
✓ Nomad Introduction
❏ Future Plan
Nomad future plans
● Develop storage and network related features if there are volunteers
that are interested.
● Try to make Nomad less VM centric
○ consider FPGA that is sitting out on PCI as an independent
resource for anything that may be deployable on it. Like VMs on
CPUs
○ Good for portability and scalability
● Work with existing teams (e.g. EPA)
● Fault management
Outline
✓ Motivation
✓ Interesting Use Cases
✓ OPNFV Requirements
✓ Nomad Introduction
✓ Future Plan
BoF Discussion
https://etherpad.openstack.org/p/AUS-BoF-NOMAD
The missing piece and Ironic
When it comes to NFV and the use of VNFs that have been created and are
managed in this way, they are not visible to Neutron and not captured in Nova as
consumable functions.
Ironic has the role of discovering and initializing “bare metal” devices and
exposing them to the rest of the OpenStack system. However there is no
requirement that all resources used by Nova for example need be bare metal
based.
We need a public api that allows the dynamic registration of resources that
happen to be hosted on acceleration hardware.
Backup Slides
What is Application Acceleration (Harm)
Typically when you run something on alternate hardware as a subcomponent of the application, we call it an
accelerated application, and the specific function is often call an accelerated function. This is commonly
called the “offload” model. Sometimes referred to as co-processing, and we see analytics as an example of
this type of workload.
The “inline” model frequently puts a specialized accelerator between an application and an interface to other
systems. We all know about Graphics “acceleration” as an example, or wetware interface pre-processors
used in genomics. An inline function may be standalone and not have any external processing dependencies
A few platforms have emerged to support this, namely GPU and FPGA, and along with even more specific
hardware are commonly connected to a more traditional CPU via PCI or similar technology.
As these patterns and specific workloads have become highly popular we have seen general CPU vendors
add acceleration platforms to the chipset. Graphics, communications, encryption are all examples.
So how do we deploy and manage acceleration hardware? and
How do we manage them separately from the service that exploites them?
OpenStack by principal - Nova (Harm)
Nova by definition manages the allocation of compute resources.
● Through meta data it is dynamically aware of a compute node and its
characteristics. This may include some close or nearby resources.
● Nova can be taught with alternate meta data about what look like standalone
compute nodes that are in fact acceleration devices.
● By providing additional filters and automation scripts Nova can manage a
standalone acceleration device just like a general CPU device
Glance is used to manage the life cycle of artifacts used for provisioning
● Glance understands the artifacts through meta data associated with a resource
● Just like VMs need images, accelerated devices need to be loaded with
bitstreams
● By providing additional artifact types and meta data Glance can be utilized to
manage the artifacts needed for acceleration device life cycle management
OpenStack by principal - Glance (Harm)
OpenStack by principal - Neutron/Cinder (Harm)
Neutron and Cinder manage traditional data center devices and do not understand
how they are implemented. The device is just something to be configured, and
separately monitored and managed.
● The devices are consumed much like a PaaS or SaaS level service. An
appliance model. If the appliance is implemented with CPU, GPU, FPGA,
discrete hardware or aliens it does not matter.
● The manager holds operational state and configuration data about the
devices, just like Nova understands number of vCPU and how much has been
allocated.
Conclusions (certainly from Harm ;-))
● This completes the puzzle and separates out the concerns for creating and
consuming VNFs to create NFV and NFVi.
● This approach provides a more generalized way to manage acceleration
hardware while still separating life cycle from specific function.
● By supporting all 3 types of configuration as well as consideration of inline
and offload models, any specific performance optimization for each can still
be applied, but without affecting the alternates.
While current steps in Nova to support acceleration, and in Nomad to support the
domain specific notions of NFV…
This approach completes the picture
OpenStack to the rescue (Harm)
By leveraging Glance and Nova to manage the provisioning of acceleration
hardware in all models, the accelerated application/function can be adopted
directly into any automation a customer needs.
Functions provided by these accelerated systems can be combined and
consumed like any PaaS or SaaS service
Not sure how to mix this in the flow (Harm)
The following set of slides provide a bit of background and then the reasoning to
lead to the following proposed approaches. Since Nomad already has a path
forward I leave it to the Nomad veterans to determine if this is of interest and how
to weave the ideas.

More Related Content

What's hot

SunGrid: Cloud Computing
SunGrid: Cloud ComputingSunGrid: Cloud Computing
SunGrid: Cloud ComputingUday Subbarayan
 
Introducing HPC with a Raspberry Pi Cluster
Introducing HPC with a Raspberry Pi ClusterIntroducing HPC with a Raspberry Pi Cluster
Introducing HPC with a Raspberry Pi Clusterinside-BigData.com
 
Using a Field Programmable Gate Array to Accelerate Application Performance
Using a Field Programmable Gate Array to Accelerate Application PerformanceUsing a Field Programmable Gate Array to Accelerate Application Performance
Using a Field Programmable Gate Array to Accelerate Application PerformanceOdinot Stanislas
 
Warehouse scale computer
Warehouse scale computerWarehouse scale computer
Warehouse scale computerHassan A-j
 
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza Databases
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza DatabasesNagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza Databases
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza DatabasesNagios
 
Red Hat Storage Day New York - What's New in Red Hat Ceph Storage
Red Hat Storage Day New York - What's New in Red Hat Ceph StorageRed Hat Storage Day New York - What's New in Red Hat Ceph Storage
Red Hat Storage Day New York - What's New in Red Hat Ceph StorageRed_Hat_Storage
 
NNSA Explorations: ARM for Supercomputing
NNSA Explorations: ARM for SupercomputingNNSA Explorations: ARM for Supercomputing
NNSA Explorations: ARM for Supercomputinginside-BigData.com
 
OpenHPC: A Comprehensive System Software Stack
OpenHPC: A Comprehensive System Software StackOpenHPC: A Comprehensive System Software Stack
OpenHPC: A Comprehensive System Software Stackinside-BigData.com
 
00 opencapi acceleration framework yonglu_ver2
00 opencapi acceleration framework yonglu_ver200 opencapi acceleration framework yonglu_ver2
00 opencapi acceleration framework yonglu_ver2Yutaka Kawai
 
Configuration management state of the art
Configuration management state of the artConfiguration management state of the art
Configuration management state of the artJean Parpaillon
 
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...Nagios
 
OpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC SystemsOpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC SystemsHPCC Systems
 
IBM Runtimes Performance Observations with Apache Spark
IBM Runtimes Performance Observations with Apache SparkIBM Runtimes Performance Observations with Apache Spark
IBM Runtimes Performance Observations with Apache SparkAdamRobertsIBM
 
A Dataflow Processing Chip for Training Deep Neural Networks
A Dataflow Processing Chip for Training Deep Neural NetworksA Dataflow Processing Chip for Training Deep Neural Networks
A Dataflow Processing Chip for Training Deep Neural Networksinside-BigData.com
 

What's hot (20)

SunGrid: Cloud Computing
SunGrid: Cloud ComputingSunGrid: Cloud Computing
SunGrid: Cloud Computing
 
Introducing HPC with a Raspberry Pi Cluster
Introducing HPC with a Raspberry Pi ClusterIntroducing HPC with a Raspberry Pi Cluster
Introducing HPC with a Raspberry Pi Cluster
 
Using a Field Programmable Gate Array to Accelerate Application Performance
Using a Field Programmable Gate Array to Accelerate Application PerformanceUsing a Field Programmable Gate Array to Accelerate Application Performance
Using a Field Programmable Gate Array to Accelerate Application Performance
 
Qct quick stack ubuntu openstack
Qct quick stack ubuntu openstackQct quick stack ubuntu openstack
Qct quick stack ubuntu openstack
 
Microblaze
MicroblazeMicroblaze
Microblaze
 
Warehouse scale computer
Warehouse scale computerWarehouse scale computer
Warehouse scale computer
 
OpenCAPI Technology Ecosystem
OpenCAPI Technology EcosystemOpenCAPI Technology Ecosystem
OpenCAPI Technology Ecosystem
 
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza Databases
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza DatabasesNagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza Databases
Nagios Conference 2014 - Frank Pantaleo - Nagios Monitoring of Netezza Databases
 
Red Hat Storage Day New York - What's New in Red Hat Ceph Storage
Red Hat Storage Day New York - What's New in Red Hat Ceph StorageRed Hat Storage Day New York - What's New in Red Hat Ceph Storage
Red Hat Storage Day New York - What's New in Red Hat Ceph Storage
 
NNSA Explorations: ARM for Supercomputing
NNSA Explorations: ARM for SupercomputingNNSA Explorations: ARM for Supercomputing
NNSA Explorations: ARM for Supercomputing
 
OpenHPC: A Comprehensive System Software Stack
OpenHPC: A Comprehensive System Software StackOpenHPC: A Comprehensive System Software Stack
OpenHPC: A Comprehensive System Software Stack
 
00 opencapi acceleration framework yonglu_ver2
00 opencapi acceleration framework yonglu_ver200 opencapi acceleration framework yonglu_ver2
00 opencapi acceleration framework yonglu_ver2
 
Configuration management state of the art
Configuration management state of the artConfiguration management state of the art
Configuration management state of the art
 
IBM HPC Transformation with AI
IBM HPC Transformation with AI IBM HPC Transformation with AI
IBM HPC Transformation with AI
 
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...
Nagios Conference 2014 - Jeremy Rust - Avoiding Downtime Using Linux High Ava...
 
NWU and HPC
NWU and HPCNWU and HPC
NWU and HPC
 
OpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC SystemsOpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC Systems
 
B9 cmis
B9 cmisB9 cmis
B9 cmis
 
IBM Runtimes Performance Observations with Apache Spark
IBM Runtimes Performance Observations with Apache SparkIBM Runtimes Performance Observations with Apache Spark
IBM Runtimes Performance Observations with Apache Spark
 
A Dataflow Processing Chip for Training Deep Neural Networks
A Dataflow Processing Chip for Training Deep Neural NetworksA Dataflow Processing Chip for Training Deep Neural Networks
A Dataflow Processing Chip for Training Deep Neural Networks
 

Viewers also liked

Serena Mah's Clinical Rotations
Serena Mah's Clinical RotationsSerena Mah's Clinical Rotations
Serena Mah's Clinical RotationsSerena Mah
 
Dieta equilibrada
Dieta equilibrada Dieta equilibrada
Dieta equilibrada LAUTOXO
 
El perfil del investigador y la difusión de la producción científica en Inter...
El perfil del investigador y la difusión de la producción científica en Inter...El perfil del investigador y la difusión de la producción científica en Inter...
El perfil del investigador y la difusión de la producción científica en Inter...Rebeca Pardo
 
Symantec™ Advanced Threat Protection: Email
Symantec™ Advanced Threat Protection: EmailSymantec™ Advanced Threat Protection: Email
Symantec™ Advanced Threat Protection: EmailSymantec Brasil
 
Hema core topics_2011.10.17
Hema core topics_2011.10.17Hema core topics_2011.10.17
Hema core topics_2011.10.17Igor Pivovarov
 
Бизнес модели 2016.03.02
Бизнес модели 2016.03.02Бизнес модели 2016.03.02
Бизнес модели 2016.03.02Igor Pivovarov
 

Viewers also liked (10)

TABLA
TABLATABLA
TABLA
 
Best Seo Pros
Best Seo ProsBest Seo Pros
Best Seo Pros
 
Serena Mah's Clinical Rotations
Serena Mah's Clinical RotationsSerena Mah's Clinical Rotations
Serena Mah's Clinical Rotations
 
антикризисные программы
антикризисные программыантикризисные программы
антикризисные программы
 
Barrall
BarrallBarrall
Barrall
 
Dieta equilibrada
Dieta equilibrada Dieta equilibrada
Dieta equilibrada
 
El perfil del investigador y la difusión de la producción científica en Inter...
El perfil del investigador y la difusión de la producción científica en Inter...El perfil del investigador y la difusión de la producción científica en Inter...
El perfil del investigador y la difusión de la producción científica en Inter...
 
Symantec™ Advanced Threat Protection: Email
Symantec™ Advanced Threat Protection: EmailSymantec™ Advanced Threat Protection: Email
Symantec™ Advanced Threat Protection: Email
 
Hema core topics_2011.10.17
Hema core topics_2011.10.17Hema core topics_2011.10.17
Hema core topics_2011.10.17
 
Бизнес модели 2016.03.02
Бизнес модели 2016.03.02Бизнес модели 2016.03.02
Бизнес модели 2016.03.02
 

Similar to Making workload nomadic when accelerated

Red hat NFV Roadmap - OpenStack Summit 2016/Red Hat NFV Mini Summit
Red hat NFV Roadmap    - OpenStack Summit 2016/Red Hat NFV Mini SummitRed hat NFV Roadmap    - OpenStack Summit 2016/Red Hat NFV Mini Summit
Red hat NFV Roadmap - OpenStack Summit 2016/Red Hat NFV Mini Summitkimw001
 
Benchmarking sahara based big data as a service solutions
Benchmarking sahara based big data as a service solutionsBenchmarking sahara based big data as a service solutions
Benchmarking sahara based big data as a service solutionsZhidong Yu
 
Stacks and Layers: Integrating P4, C, OVS and OpenStack
Stacks and Layers: Integrating P4, C, OVS and OpenStackStacks and Layers: Integrating P4, C, OVS and OpenStack
Stacks and Layers: Integrating P4, C, OVS and OpenStackOpen-NFP
 
SDN and NFV Friends or Enemies ?
SDN and NFV Friends or Enemies ?SDN and NFV Friends or Enemies ?
SDN and NFV Friends or Enemies ?Kedar Raval
 
Know about SDN and NFV
Know about SDN and NFVKnow about SDN and NFV
Know about SDN and NFVKedar Raval
 
Seminar Accelerating Business Using Microservices Architecture in Digital Age...
Seminar Accelerating Business Using Microservices Architecture in Digital Age...Seminar Accelerating Business Using Microservices Architecture in Digital Age...
Seminar Accelerating Business Using Microservices Architecture in Digital Age...PT Datacomm Diangraha
 
Technical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Technical Report NetApp Clustered Data ONTAP 8.2: An IntroductionTechnical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Technical Report NetApp Clustered Data ONTAP 8.2: An IntroductionNetApp
 
NFV Orchestration for Optimal Performance
NFV Orchestration for Optimal PerformanceNFV Orchestration for Optimal Performance
NFV Orchestration for Optimal Performancedfilppi
 
Challenges in positioning open stack for nf-vi_ are we biting off more than w...
Challenges in positioning open stack for nf-vi_ are we biting off more than w...Challenges in positioning open stack for nf-vi_ are we biting off more than w...
Challenges in positioning open stack for nf-vi_ are we biting off more than w...OPNFV
 
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...Sadique Puthen
 
Lookout on Scaling Security to 100 Million Devices
Lookout on Scaling Security to 100 Million DevicesLookout on Scaling Security to 100 Million Devices
Lookout on Scaling Security to 100 Million DevicesScyllaDB
 
Red Hat presentatie: Open stack Latest Pure Tech
Red Hat presentatie: Open stack Latest Pure TechRed Hat presentatie: Open stack Latest Pure Tech
Red Hat presentatie: Open stack Latest Pure TechProxyServices
 
SDN and NFV: Friends or Enemies
SDN and NFV: Friends or EnemiesSDN and NFV: Friends or Enemies
SDN and NFV: Friends or EnemiesJustyna Bak
 
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guide
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guideBasic concepts for_clustered_data_ontap_8.3_v1.1-lab_guide
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guideVikas Sharma
 
PFQ@ 9th Italian Networking Workshop (Courmayeur)
PFQ@ 9th Italian Networking Workshop (Courmayeur)PFQ@ 9th Italian Networking Workshop (Courmayeur)
PFQ@ 9th Italian Networking Workshop (Courmayeur)Nicola Bonelli
 
OpenStack Paris Meetup on Nfv 2014/10/07
OpenStack Paris Meetup on Nfv 2014/10/07OpenStack Paris Meetup on Nfv 2014/10/07
OpenStack Paris Meetup on Nfv 2014/10/07Nicolas (Nick) Barcet
 
OpenStack Toronto: Juno Community Update
OpenStack Toronto: Juno Community UpdateOpenStack Toronto: Juno Community Update
OpenStack Toronto: Juno Community UpdateStephen Gordon
 
OpenStack Best Practices and Considerations - terasky tech day
OpenStack Best Practices and Considerations  - terasky tech dayOpenStack Best Practices and Considerations  - terasky tech day
OpenStack Best Practices and Considerations - terasky tech dayArthur Berezin
 

Similar to Making workload nomadic when accelerated (20)

Red hat NFV Roadmap - OpenStack Summit 2016/Red Hat NFV Mini Summit
Red hat NFV Roadmap    - OpenStack Summit 2016/Red Hat NFV Mini SummitRed hat NFV Roadmap    - OpenStack Summit 2016/Red Hat NFV Mini Summit
Red hat NFV Roadmap - OpenStack Summit 2016/Red Hat NFV Mini Summit
 
Benchmarking sahara based big data as a service solutions
Benchmarking sahara based big data as a service solutionsBenchmarking sahara based big data as a service solutions
Benchmarking sahara based big data as a service solutions
 
Stacks and Layers: Integrating P4, C, OVS and OpenStack
Stacks and Layers: Integrating P4, C, OVS and OpenStackStacks and Layers: Integrating P4, C, OVS and OpenStack
Stacks and Layers: Integrating P4, C, OVS and OpenStack
 
SDN and NFV Friends or Enemies ?
SDN and NFV Friends or Enemies ?SDN and NFV Friends or Enemies ?
SDN and NFV Friends or Enemies ?
 
Know about SDN and NFV
Know about SDN and NFVKnow about SDN and NFV
Know about SDN and NFV
 
Cloud Networking Trends
Cloud Networking TrendsCloud Networking Trends
Cloud Networking Trends
 
Seminar Accelerating Business Using Microservices Architecture in Digital Age...
Seminar Accelerating Business Using Microservices Architecture in Digital Age...Seminar Accelerating Business Using Microservices Architecture in Digital Age...
Seminar Accelerating Business Using Microservices Architecture in Digital Age...
 
Technical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Technical Report NetApp Clustered Data ONTAP 8.2: An IntroductionTechnical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Technical Report NetApp Clustered Data ONTAP 8.2: An Introduction
 
NFV Orchestration for Optimal Performance
NFV Orchestration for Optimal PerformanceNFV Orchestration for Optimal Performance
NFV Orchestration for Optimal Performance
 
Challenges in positioning open stack for nf-vi_ are we biting off more than w...
Challenges in positioning open stack for nf-vi_ are we biting off more than w...Challenges in positioning open stack for nf-vi_ are we biting off more than w...
Challenges in positioning open stack for nf-vi_ are we biting off more than w...
 
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...
Challenges in Positioning OpenStack for NFVi: Are We Biting Off More Than We ...
 
Lookout on Scaling Security to 100 Million Devices
Lookout on Scaling Security to 100 Million DevicesLookout on Scaling Security to 100 Million Devices
Lookout on Scaling Security to 100 Million Devices
 
Red Hat presentatie: Open stack Latest Pure Tech
Red Hat presentatie: Open stack Latest Pure TechRed Hat presentatie: Open stack Latest Pure Tech
Red Hat presentatie: Open stack Latest Pure Tech
 
SDN and NFV: Friends or Enemies
SDN and NFV: Friends or EnemiesSDN and NFV: Friends or Enemies
SDN and NFV: Friends or Enemies
 
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guide
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guideBasic concepts for_clustered_data_ontap_8.3_v1.1-lab_guide
Basic concepts for_clustered_data_ontap_8.3_v1.1-lab_guide
 
PFQ@ 9th Italian Networking Workshop (Courmayeur)
PFQ@ 9th Italian Networking Workshop (Courmayeur)PFQ@ 9th Italian Networking Workshop (Courmayeur)
PFQ@ 9th Italian Networking Workshop (Courmayeur)
 
OpenStack Paris Meetup on Nfv 2014/10/07
OpenStack Paris Meetup on Nfv 2014/10/07OpenStack Paris Meetup on Nfv 2014/10/07
OpenStack Paris Meetup on Nfv 2014/10/07
 
OpenStack Toronto: Juno Community Update
OpenStack Toronto: Juno Community UpdateOpenStack Toronto: Juno Community Update
OpenStack Toronto: Juno Community Update
 
OpenStack Best Practices and Considerations - terasky tech day
OpenStack Best Practices and Considerations  - terasky tech dayOpenStack Best Practices and Considerations  - terasky tech day
OpenStack Best Practices and Considerations - terasky tech day
 
Bbva bank on Open Stack
Bbva bank on Open StackBbva bank on Open Stack
Bbva bank on Open Stack
 

Recently uploaded

Modern Roaming for Notes and Nomad – Cheaper Faster Better Stronger
Modern Roaming for Notes and Nomad – Cheaper Faster Better StrongerModern Roaming for Notes and Nomad – Cheaper Faster Better Stronger
Modern Roaming for Notes and Nomad – Cheaper Faster Better Strongerpanagenda
 
Manual 508 Accessibility Compliance Audit
Manual 508 Accessibility Compliance AuditManual 508 Accessibility Compliance Audit
Manual 508 Accessibility Compliance AuditSkynet Technologies
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .Alan Dix
 
Generative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersGenerative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersRaghuram Pandurangan
 
Moving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfMoving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfLoriGlavin3
 
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024BookNet Canada
 
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Alkin Tezuysal
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...Wes McKinney
 
Testing tools and AI - ideas what to try with some tool examples
Testing tools and AI - ideas what to try with some tool examplesTesting tools and AI - ideas what to try with some tool examples
Testing tools and AI - ideas what to try with some tool examplesKari Kakkonen
 
Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Hiroshi SHIBATA
 
Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationKnoldus Inc.
 
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...panagenda
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsRavi Sanghani
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch TuesdayIvanti
 
Generative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfGenerative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfIngrid Airi González
 
Assure Ecommerce and Retail Operations Uptime with ThousandEyes
Assure Ecommerce and Retail Operations Uptime with ThousandEyesAssure Ecommerce and Retail Operations Uptime with ThousandEyes
Assure Ecommerce and Retail Operations Uptime with ThousandEyesThousandEyes
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxLoriGlavin3
 
Sample pptx for embedding into website for demo
Sample pptx for embedding into website for demoSample pptx for embedding into website for demo
Sample pptx for embedding into website for demoHarshalMandlekar2
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfNeo4j
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxLoriGlavin3
 

Recently uploaded (20)

Modern Roaming for Notes and Nomad – Cheaper Faster Better Stronger
Modern Roaming for Notes and Nomad – Cheaper Faster Better StrongerModern Roaming for Notes and Nomad – Cheaper Faster Better Stronger
Modern Roaming for Notes and Nomad – Cheaper Faster Better Stronger
 
Manual 508 Accessibility Compliance Audit
Manual 508 Accessibility Compliance AuditManual 508 Accessibility Compliance Audit
Manual 508 Accessibility Compliance Audit
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .
 
Generative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersGenerative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information Developers
 
Moving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfMoving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdf
 
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: Loan Stars - Tech Forum 2024
 
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
 
Testing tools and AI - ideas what to try with some tool examples
Testing tools and AI - ideas what to try with some tool examplesTesting tools and AI - ideas what to try with some tool examples
Testing tools and AI - ideas what to try with some tool examples
 
Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024
 
Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog Presentation
 
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...
Why device, WIFI, and ISP insights are crucial to supporting remote Microsoft...
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and Insights
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch Tuesday
 
Generative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfGenerative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdf
 
Assure Ecommerce and Retail Operations Uptime with ThousandEyes
Assure Ecommerce and Retail Operations Uptime with ThousandEyesAssure Ecommerce and Retail Operations Uptime with ThousandEyes
Assure Ecommerce and Retail Operations Uptime with ThousandEyes
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
 
Sample pptx for embedding into website for demo
Sample pptx for embedding into website for demoSample pptx for embedding into website for demo
Sample pptx for embedding into website for demo
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdf
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
 

Making workload nomadic when accelerated

  • 1. Making Workload Nomadic When Accelerated Nomad Project Introduction Zhipeng Huang - Huawei Michele Paolino - Virtual Open Systems Hui Deng - China Mobile
  • 2. Outline ❏ Motivation ❏ Interesting Use Cases ❏ OPNFV Requirements ❏ Nomad Introduction ❏ Future Plan
  • 4. Motivation - Why ? OpenStack currently lacks of a set of features in respect to portable hardware acceleration*: ● Accelerators life management ● Accelerators resource discovery ● Reconfigurable FPGA, GPUs and other accelerators migration support, ease of use, etc. * these features have been highlighted in the OPNFV OpenStack GAP Analysis document. The main purpose of Nomad is to implement all these features, providing (all) the OpenStack users with a portable, powerful, easy way to use hardware acceleration
  • 5. Motivation - What ? 3 models if related to a more traditional CPU as part of the workload Very Close - in the cpu chipset or on the board (i.e., see Intel’s skylake staging) ● Suitable for offload model, and inline if associated interface is also in place ● With optimal sharing of resources can provide excellent processing gains ● Limited by horizontal scale, but can be leveraged as a unit of management like the associated CPU Nearby - attached via a bus or similar (i.e., PCIe or within a chassis assembly) ● Suitable for offload and inline models ● Susceptible to negative impact if interface across the bus is chatty ● Larger scale possible, particularly in chassis configurations Far - reachable by TCP/IP or other communication protocol ● Suitable for offload and inline if latency is not a concern ● Largest horizontal scale flexibility ● Much more suited to a standalone function model
  • 6. Motivation - How ? Nova Extension Solely Nomad (Dedicated Management Function) Target Best performance, no portability. Best performance/portability trade off. Accelerator access Direct management. Management through a portability layer. Pros - Direct interaction between the compute node and the accelerators could provide slightly better performance - Resource discovery, scheduling, setup etc. - Support for automatic integrated acceleration management for accelerated VM's Migration - Hardware portability and independence. Cons - Portability/migration hard to support - Code complexity: specific code needed for each accelerator type, with impact to the project performance, security, maintainability, etc. - The accelerator allocation phase might take time, as an handshake procedure has to be put in place - Scalability issues
  • 7. Outline ✓ Motivation ❏ Interesting Use Cases ❏ OPNFV Requirements ❏ Nomad Introduction ❏ Future Plan
  • 8. Interesting Use Cases - NFVIaaS ● For people that are familiar with ETSI NFV standard, NFVIaaS was among the NFV use cases that published in Phase 1 documents ● However few of us grasp what this use case actually meant for business, until now. ● Now many operators begin to build their own Public Cloud. If an operator has a multiple site supported Public Cloud, it then would be able to offer NFVIaaS to NFV app companies that has its own content, or VNF, or MANO, but without NFVI. These NFV apps then could be deployed on the Public Cloud. ● There is still a problem for the operator who owns the Public Cloud - so further service classification on NFVIaaS without acceleration Acceleration makes NFVIaaS a possibility
  • 9. Interesting Use Cases - accelerated vSwitch The next generation of virtual switch technologies will combine data plane and low level SW acceleration with network engines, PCI accelerators, etc: ● Dynamically allocating (reconfigurable) accelerators and supporting VM migration ● Dispatching packet processing workloads on the available resources *http://www.virtualopensystems.com/en/products/vosyswitch-nfv-virtual-switch/ VOSYSwitch* is a user space virtual switch for NFV (and not only), based on the framework Snabb (snabb.io), and thus independent from Intel/DPDK: ● It inherits from Snabb the LuaJIT trace compiler, which provides performance optimization at runtime based on the current traffic ● Supports x86/ARM architectures and hardware accelerators through ODP ● Its roadmap includes FPGA/GPU/OpenCL support, and integration with OpenStack Nomad
  • 10. Outline ✓ Motivation ✓ Interesting Use Cases ❏ OPNFV Requirements ❏ Nomad Introduction ❏ Future Plan
  • 11. OPNFV Requirements DPACC general requirments : https://docs. google. com/document/d/1YexfnL RZ9gZnj- 5PNOnrJ5CVMhbzZ7mIJo gW6nyIGq0/edit DPACC OpenStack requirements : https: //docs.google. com/document/d/1_fOinIQ NcPwNODZPzGK5vRMPJ QLwL7iLds4NFnjXSms/ed it
  • 12. Outline ✓ Motivation ✓ Interesting Use Cases ✓ OPNFV Requirements ❏ Nomad Introduction ❏ Future Plan
  • 13. Nomad Introduction - some really high level stuff Design Goal: ● Loosely coupled (more than Neutron) ● DB oriented management ● Network: DAG based ● Storage: affinity driven ● Compute : capability based scheduling ● Big Data: FPGA/GPU programming ● Fine grained fault monitoring and management (bloom filter?)
  • 14. Nomad introduction - first stab at nomad-compute ● BP Driven, everything starts from scratch ● Target code available in August ● Try to make OPNFV Colorado release if possible
  • 15. Outline ✓ Motivation ✓ Interesting Use Cases ✓ OPNFV Requirements ✓ Nomad Introduction ❏ Future Plan
  • 16. Nomad future plans ● Develop storage and network related features if there are volunteers that are interested. ● Try to make Nomad less VM centric ○ consider FPGA that is sitting out on PCI as an independent resource for anything that may be deployable on it. Like VMs on CPUs ○ Good for portability and scalability ● Work with existing teams (e.g. EPA) ● Fault management
  • 17. Outline ✓ Motivation ✓ Interesting Use Cases ✓ OPNFV Requirements ✓ Nomad Introduction ✓ Future Plan
  • 19. The missing piece and Ironic When it comes to NFV and the use of VNFs that have been created and are managed in this way, they are not visible to Neutron and not captured in Nova as consumable functions. Ironic has the role of discovering and initializing “bare metal” devices and exposing them to the rest of the OpenStack system. However there is no requirement that all resources used by Nova for example need be bare metal based. We need a public api that allows the dynamic registration of resources that happen to be hosted on acceleration hardware.
  • 21. What is Application Acceleration (Harm) Typically when you run something on alternate hardware as a subcomponent of the application, we call it an accelerated application, and the specific function is often call an accelerated function. This is commonly called the “offload” model. Sometimes referred to as co-processing, and we see analytics as an example of this type of workload. The “inline” model frequently puts a specialized accelerator between an application and an interface to other systems. We all know about Graphics “acceleration” as an example, or wetware interface pre-processors used in genomics. An inline function may be standalone and not have any external processing dependencies A few platforms have emerged to support this, namely GPU and FPGA, and along with even more specific hardware are commonly connected to a more traditional CPU via PCI or similar technology. As these patterns and specific workloads have become highly popular we have seen general CPU vendors add acceleration platforms to the chipset. Graphics, communications, encryption are all examples. So how do we deploy and manage acceleration hardware? and How do we manage them separately from the service that exploites them?
  • 22. OpenStack by principal - Nova (Harm) Nova by definition manages the allocation of compute resources. ● Through meta data it is dynamically aware of a compute node and its characteristics. This may include some close or nearby resources. ● Nova can be taught with alternate meta data about what look like standalone compute nodes that are in fact acceleration devices. ● By providing additional filters and automation scripts Nova can manage a standalone acceleration device just like a general CPU device
  • 23. Glance is used to manage the life cycle of artifacts used for provisioning ● Glance understands the artifacts through meta data associated with a resource ● Just like VMs need images, accelerated devices need to be loaded with bitstreams ● By providing additional artifact types and meta data Glance can be utilized to manage the artifacts needed for acceleration device life cycle management OpenStack by principal - Glance (Harm)
  • 24. OpenStack by principal - Neutron/Cinder (Harm) Neutron and Cinder manage traditional data center devices and do not understand how they are implemented. The device is just something to be configured, and separately monitored and managed. ● The devices are consumed much like a PaaS or SaaS level service. An appliance model. If the appliance is implemented with CPU, GPU, FPGA, discrete hardware or aliens it does not matter. ● The manager holds operational state and configuration data about the devices, just like Nova understands number of vCPU and how much has been allocated.
  • 25. Conclusions (certainly from Harm ;-)) ● This completes the puzzle and separates out the concerns for creating and consuming VNFs to create NFV and NFVi. ● This approach provides a more generalized way to manage acceleration hardware while still separating life cycle from specific function. ● By supporting all 3 types of configuration as well as consideration of inline and offload models, any specific performance optimization for each can still be applied, but without affecting the alternates. While current steps in Nova to support acceleration, and in Nomad to support the domain specific notions of NFV… This approach completes the picture
  • 26. OpenStack to the rescue (Harm) By leveraging Glance and Nova to manage the provisioning of acceleration hardware in all models, the accelerated application/function can be adopted directly into any automation a customer needs. Functions provided by these accelerated systems can be combined and consumed like any PaaS or SaaS service
  • 27. Not sure how to mix this in the flow (Harm) The following set of slides provide a bit of background and then the reasoning to lead to the following proposed approaches. Since Nomad already has a path forward I leave it to the Nomad veterans to determine if this is of interest and how to weave the ideas.