Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Stokes, Ian]

Hi all,

Just an update on below, I'm beginning to create a document for submission to 
the OVS repo to capture the work items identified in the google doc.

I'm looking to submit this next week (probably Wednesday the 13th). In the 
meantime can I ask anyone interested in listing a feature they are working on 
for the 2.9 release to please add it to the google doc.

https://docs.google.com/spreadsheets/d/1FilGq46vQePFKoehADWWsDvDCZSNsMU7PrpPr3EM3lU/edit#gid=0

Note: It is not mandatory to add features that are being worked on to this 
document but it is encouraged. If a feature is not listed here but is submitted 
there is still every chance it will make it into the 2.9 release.

The goal of this document is simply to improve visibility of ongoing work in 
within the community for upcoming releases.

Thanks
Ian

> Below are the features that Intel are planning to submit for the OVS 2.9
> release.
> 
> I've also created a google spreadsheet so that people can contribute their
> own planned features. This can be found at:
> 
> https://docs.google.com/spreadsheets/d/1FilGq46vQePFKoehADWWsDvDCZSNsMU7Pr
> pPr3EM3lU/edit#gid=0
> 
> If people can complete this doc and send an email to the dev mailing list
> then we can formalize the features and commit them to a doc in the OVS
> repo.
> 
> It would be good if others are also willing to share their plans so that
> we can build up a complete picture of what's planned for the OVS 2.9
> release and make sure there's no duplication.
> 
> Keepalive: Keepalive feature is aimed at achieving Fastpath Service
> Assurance  in OVS-DPDK deployments. It adds support for monitoring the
> packet processing threads by dispatching heartbeats at regular intervals
> and the status is updated periodically in OVSDB. In case of heartbeat
> misses the failure is detected and reported to higher level fault
> management systems/frameworks. The status can be monitored from OpenStack
> Ceilometer service.
> 
> vHost PMD: The vHost PMD brings vHost User ports under control of DPDK's
> librte_ether library and in doing so DPDK's librte_vhost library will no
> longer be directly referenced in OVS. Potential benefits include shared
> code among port types in netdev-dpdk and smoother upgrades between DPDK
> versions in OVS.
> 
> Multi-segment Mbuf: Support for jumbo frames was added to netdev-dpdk in
> v2.6.0. In that implementation, a jumbo frame is stored within a single-
> segment mbuf. That mode will remain the default, but an option will now be
> added to allow a jumbo frame to be stored within a multi-segment mbuf.
> This reduces the requirement for large, contiguous blocks of memory, and
> may be useful in deployments with limited memory.
> 
> Conn Track: Analyze and improve Connection Tracker performance:
> The Connection Tracker is a feature to manage stateful connections and
> implement security Firewalls.
> This allows a better protection against attacks and helps in load
> balancing. The counterpart is a significant impact on the overall
> performance. This work is aimed at analyzing possible bottlenecks - also
> considering the latest protocol implementations - to improve the
> Connection Tracker performance.
> 
> IPSEC: This feature looks to introduce IPsec into OVS with DPDK. IPsec
> would function in transport mode and would be used in conjunction with
> existing encapsulation methods (initially VxLAN) to create a new interface
> type 'vxlanipsec'. The DPDK cryptodev framework will be used to  handle
> cipher/digest operations as part of the encap/decap actions in accordance
> with a security association. The cryptodev devices supported would be
> limited to virtual crypto devices such as the  AESN-MB vdev. As such
> cipher and digest algorithms supported would be limited to the
> capabilities of the vdev. In terms of Security Association generation for
> IPsec policies, the feature will allow a user to specify security
> associations via command line options for the interface.
> 
> Zero copy: Support for enabling DPDK's 'dequeue zero copy' feature on
> vHost User ports which removes the costly memcpy operation from guest to
> host when enabled. Detailed information in the DPDK documentation:
> http://dpdk.org/doc/guides/prog_guide/vhost_lib.html#vhost-api-overview.
> 
> DPDK 17.05.2 and 17.11 support: DPDK 17.05.2 is the latest stable release
> of DPDK, and includes numerous bug-fixes and stability updates.
> DPDK 17.11 will be the next DPDK Long Term Support (LTS) release (TBD). It
> will also include numerous new features, such as Generic Segmentation
> Offload (GSO) and vHost PMD.
> 
> Upcall performance: The first packet in every flow that is handled by OVS
> is an upcall to the exception path. This is a performance bottleneck for
> some use cases. Analysis and improvements to the performance of this code
> path will be investigated.
> 
> Service Assurance : Virtual switching reporting of flow telemetry via
> IPFix.
> 
> Extended NIC Stats: Expose low level driver 

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Stokes, Ian]

> 
> That is the one in flight now.
> I guess there are no others being planned, then ?
> 

Not from Intel that I am aware of.

> 
> -Original Message-
> From: "Stokes, Ian" <ian.sto...@intel.com>
> Date: Monday, August 14, 2017 at 2:56 AM
> To: Jan Scheurich <jan.scheur...@ericsson.com>, Darrell Ball
> <db...@vmware.com>, "ovs-dev@openvswitch.org" <ovs-dev@openvswitch.org>
> Cc: "O Mahony, Billy" <billy.o.mah...@intel.com>
> Subject: RE: [ovs-dev] OVS 2.9 Intel Roadmap
> 
> : RE: [ovs-dev] OVS 2.9 Intel Roadmap
> 
> >
> 
> > What about Billy's patch set for Rx queue prioritization?
> 
> > https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__mail.openvswitch.org_pipermail_ovs-2Ddev_2017-
> 2DJuly_336001.html=DwIGaQ=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-
> uZnsw=1BYS5DJYkjWJD4seuUiAzwg-
> 7v6sQdRxBkD4MCGY45Y=ZAVlg8xWiiYtEprqO1iX0DbCTyJK3wiU86HTM5p1CoM=
> 
> > Jan
> 
> >
> 
> Good catch Jan, I missed that myself (Billy was out of office last
> week when putting the list together). I'll add this to the list.
> 
> 
> 
> Thanks
> 
> Ian
> 
> > > -Original Message-
> 
> > > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> 
>     > > boun...@openvswitch.org] On Behalf Of Stokes, Ian
> 
> > > Sent: Monday, 14 August, 2017 10:45
> 
> > > To: Darrell Ball <db...@vmware.com>; ovs-dev@openvswitch.org
> 
> > > Subject: Re: [ovs-dev] OVS 2.9 Intel Roadmap
> 
> > >
> 
> > > > Any prioritization or other qos features missing ?
> 
> > > >
> 
> > > Hi Darrell,
> 
> > >
> 
> > > Do you mean the 2 qos patchsets on the list currently (support for
> 
> > > queues and move qos in do_tx_copy?).
> 
> > >
> 
> > > If so, I didn’t include these in the list as Intel were not the
> 
> > > authors. I do plan to provide feedback on them though but the last
> few
> 
> > > weeks have been a bit hectic on my side.
> 
> > >
> 
> > > Are there any other QoS features you are referring to that are
> missing?
> 
> > >
> 
> > > Ian
> 
> > > ___
> 
> > > dev mailing list
> 
> > > d...@openvswitch.org
> 
> > > https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__mail.openvswitch.org_mailman_listinfo_ovs-
> 2Ddev=DwIGaQ=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-
> uZnsw=1BYS5DJYkjWJD4seuUiAzwg-
> 7v6sQdRxBkD4MCGY45Y=AQShd_s5gqoCXl19kdh_pwRw7qYPflOwltla-XoZVgI=
> 
> 

___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Darrell Ball]

That is the one in flight now.
I guess there are no others being planned, then ?


-Original Message-
From: "Stokes, Ian" <ian.sto...@intel.com>
Date: Monday, August 14, 2017 at 2:56 AM
To: Jan Scheurich <jan.scheur...@ericsson.com>, Darrell Ball 
<db...@vmware.com>, "ovs-dev@openvswitch.org" <ovs-dev@openvswitch.org>
Cc: "O Mahony, Billy" <billy.o.mah...@intel.com>
Subject: RE: [ovs-dev] OVS 2.9 Intel Roadmap

: RE: [ovs-dev] OVS 2.9 Intel Roadmap

> 

> What about Billy's patch set for Rx queue prioritization?

> 
https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.openvswitch.org_pipermail_ovs-2Ddev_2017-2DJuly_336001.html=DwIGaQ=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-uZnsw=1BYS5DJYkjWJD4seuUiAzwg-7v6sQdRxBkD4MCGY45Y=ZAVlg8xWiiYtEprqO1iX0DbCTyJK3wiU86HTM5p1CoM=
 

> Jan

> 

Good catch Jan, I missed that myself (Billy was out of office last week 
when putting the list together). I'll add this to the list.



Thanks

Ian

> > -Original Message-

> > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-

> > boun...@openvswitch.org] On Behalf Of Stokes, Ian

> > Sent: Monday, 14 August, 2017 10:45
    
    > > To: Darrell Ball <db...@vmware.com>; ovs-dev@openvswitch.org

> > Subject: Re: [ovs-dev] OVS 2.9 Intel Roadmap

> >

> > > Any prioritization or other qos features missing ?

> > >

> > Hi Darrell,

> >

> > Do you mean the 2 qos patchsets on the list currently (support for

> > queues and move qos in do_tx_copy?).

> >

> > If so, I didn’t include these in the list as Intel were not the

> > authors. I do plan to provide feedback on them though but the last few

> > weeks have been a bit hectic on my side.

> >

> > Are there any other QoS features you are referring to that are missing?

> >

> > Ian

> > ___

> > dev mailing list

> > d...@openvswitch.org

> > 
https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.openvswitch.org_mailman_listinfo_ovs-2Ddev=DwIGaQ=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-uZnsw=1BYS5DJYkjWJD4seuUiAzwg-7v6sQdRxBkD4MCGY45Y=AQShd_s5gqoCXl19kdh_pwRw7qYPflOwltla-XoZVgI=
 



___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Stokes, Ian]

: RE: [ovs-dev] OVS 2.9 Intel Roadmap
> 
> What about Billy's patch set for Rx queue prioritization?
> https://mail.openvswitch.org/pipermail/ovs-dev/2017-July/336001.html
> Jan
> 
Good catch Jan, I missed that myself (Billy was out of office last week when 
putting the list together). I'll add this to the list.

Thanks
Ian
> > -Original Message-
> > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> > boun...@openvswitch.org] On Behalf Of Stokes, Ian
> > Sent: Monday, 14 August, 2017 10:45
> > To: Darrell Ball <db...@vmware.com>; ovs-dev@openvswitch.org
> > Subject: Re: [ovs-dev] OVS 2.9 Intel Roadmap
> >
> > > Any prioritization or other qos features missing ?
> > >
> > Hi Darrell,
> >
> > Do you mean the 2 qos patchsets on the list currently (support for
> > queues and move qos in do_tx_copy?).
> >
> > If so, I didn’t include these in the list as Intel were not the
> > authors. I do plan to provide feedback on them though but the last few
> > weeks have been a bit hectic on my side.
> >
> > Are there any other QoS features you are referring to that are missing?
> >
> > Ian
> > ___
> > dev mailing list
> > d...@openvswitch.org
> > https://mail.openvswitch.org/mailman/listinfo/ovs-dev
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Jan Scheurich]

What about Billy's patch set for Rx queue prioritization?
https://mail.openvswitch.org/pipermail/ovs-dev/2017-July/336001.html
Jan

> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Stokes, Ian
> Sent: Monday, 14 August, 2017 10:45
> To: Darrell Ball <db...@vmware.com>; ovs-dev@openvswitch.org
> Subject: Re: [ovs-dev] OVS 2.9 Intel Roadmap
> 
> > Any prioritization or other qos features missing ?
> >
> Hi Darrell,
> 
> Do you mean the 2 qos patchsets on the list currently (support for queues
> and move qos in do_tx_copy?).
> 
> If so, I didn’t include these in the list as Intel were not the authors. I do 
> plan
> to provide feedback on them though but the last few weeks have been a bit
> hectic on my side.
> 
> Are there any other QoS features you are referring to that are missing?
> 
> Ian
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Stokes, Ian]

> Any prioritization or other qos features missing ?
> 
Hi Darrell,

Do you mean the 2 qos patchsets on the list currently (support for queues and 
move qos in do_tx_copy?).

If so, I didn’t include these in the list as Intel were not the authors. I do 
plan to provide feedback on them though but the last few weeks have been a bit 
hectic on my side.

Are there any other QoS features you are referring to that are missing?

Ian
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Darrell Ball]

Any prioritization or other qos features missing ?


-Original Message-
From: <ovs-dev-boun...@openvswitch.org> on behalf of "Stokes, Ian" 
<ian.sto...@intel.com>
Date: Friday, August 11, 2017 at 8:54 AM
To: "ovs-dev@openvswitch.org" <ovs-dev@openvswitch.org>
Subject: [ovs-dev] OVS 2.9 Intel Roadmap

Below are the features that Intel are planning to submit for the OVS 2.9 
release.

I've also created a google spreadsheet so that people can contribute their 
own planned features. This can be found at:


https://urldefense.proofpoint.com/v2/url?u=https-3A__docs.google.com_spreadsheets_d_1FilGq46vQePFKoehADWWsDvDCZSNsMU7PrpPr3EM3lU_edit-23gid-3D0=DwICAg=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-uZnsw=nnRkfhFw7IvUrJlAF8STuyl-RP9HOYU1qV2SkuscjuY=zNu3CkwzL63RcmiOcqXe9xBWKKs-wmNaqOOMeuekThY=
 

If people can complete this doc and send an email to the dev mailing list 
then we can formalize the features and commit them to a doc in the OVS repo.

It would be good if others are also willing to share their plans so that we 
can build up a complete picture of what's planned for the OVS 2.9 release and 
make sure there's no duplication.

Keepalive: Keepalive feature is aimed at achieving Fastpath Service 
Assurance  in OVS-DPDK deployments. It adds support for monitoring the packet 
processing threads by dispatching heartbeats at regular intervals and the 
status is updated periodically in OVSDB. In case of heartbeat misses the 
failure is detected and reported to higher level fault management 
systems/frameworks. The status can be monitored from OpenStack Ceilometer 
service.

vHost PMD: The vHost PMD brings vHost User ports under control of DPDK's 
librte_ether library and in doing so DPDK's librte_vhost library will no longer 
be directly referenced in OVS. Potential benefits include shared code among 
port types in netdev-dpdk and smoother upgrades between DPDK versions in OVS.

Multi-segment Mbuf: Support for jumbo frames was added to netdev-dpdk in 
v2.6.0. In that implementation, a jumbo frame is stored within a single-segment 
mbuf. That mode will remain the default, but an option will now be added to 
allow a jumbo frame to be stored within a multi-segment mbuf. This reduces the 
requirement for large, contiguous blocks of memory, and may be useful in 
deployments with limited memory.

Conn Track: Analyze and improve Connection Tracker performance:
The Connection Tracker is a feature to manage stateful connections and 
implement security Firewalls.
This allows a better protection against attacks and helps in load 
balancing. The counterpart is a significant impact on the overall performance. 
This work is aimed at analyzing possible bottlenecks - also considering the 
latest protocol implementations - to improve the Connection Tracker performance.

IPSEC: This feature looks to introduce IPsec into OVS with DPDK. IPsec 
would function in transport mode and would be used in conjunction with existing 
encapsulation methods (initially VxLAN) to create a new interface type 
'vxlanipsec'. The DPDK cryptodev framework will be used to  handle 
cipher/digest operations as part of the encap/decap actions in accordance with 
a security association. The cryptodev devices supported would be limited to 
virtual crypto devices such as the  AESN-MB vdev. As such cipher and digest 
algorithms supported would be limited to the capabilities of the vdev. In terms 
of Security Association generation for IPsec policies, the feature will allow a 
user to specify security associations via command line options for the 
interface.

Zero copy: Support for enabling DPDK's 'dequeue zero copy' feature on vHost 
User ports which removes the costly memcpy operation from guest to host when 
enabled. Detailed information in the DPDK documentation: 
https://urldefense.proofpoint.com/v2/url?u=http-3A__dpdk.org_doc_guides_prog-5Fguide_vhost-5Flib.html-23vhost-2Dapi-2Doverview=DwICAg=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-uZnsw=nnRkfhFw7IvUrJlAF8STuyl-RP9HOYU1qV2SkuscjuY=mhU8CQHE1fWvN8qYotMHZnCnMSMjsDbKBUblD2M1Gjk=
 .

DPDK 17.05.2 and 17.11 support: DPDK 17.05.2 is the latest stable release 
of DPDK, and includes numerous bug-fixes and stability updates.
DPDK 17.11 will be the next DPDK Long Term Support (LTS) release (TBD). It 
will also include numerous new features, such as Generic Segmentation Offload 
(GSO) and vHost PMD.

Upcall performance: The first packet in every flow that is handled by OVS 
is an upcall to the exception path. This is a performance bottleneck for some 
use cases. Analysis and improvements to the performance of this code path will 
be investigated.

Service Assurance : Virtual switching reporting of flow telemetry via IPFix.

Extended NIC Stats: Expose low level driver statistics registers to the 
user through the Extended NIC statistics API in DPDK

Re: [ovs-dev] OVS 2.9 Intel Roadmap

[Justin Pettit]

> On Aug 11, 2017, at 8:54 AM, Stokes, Ian  wrote:
> 
> Below are the features that Intel are planning to submit for the OVS 2.9 
> release.
> 
> I've also created a google spreadsheet so that people can contribute their 
> own planned features. This can be found at:
> 
> https://docs.google.com/spreadsheets/d/1FilGq46vQePFKoehADWWsDvDCZSNsMU7PrpPr3EM3lU/edit#gid=0
> 
> If people can complete this doc and send an email to the dev mailing list 
> then we can formalize the features and commit them to a doc in the OVS repo.
> 
> It would be good if others are also willing to share their plans so that we 
> can build up a complete picture of what's planned for the OVS 2.9 release and 
> make sure there's no duplication.

Thanks for setting this up, Ian.  I added some items that Ben and I were 
thinking that VMware will try to add.  I've asked the team to add theirs as 
well, so it may grow a bit more.

--Justin


___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] OVS 2.9 Intel Roadmap

[Stokes, Ian]

Below are the features that Intel are planning to submit for the OVS 2.9 
release.

I've also created a google spreadsheet so that people can contribute their own 
planned features. This can be found at:

https://docs.google.com/spreadsheets/d/1FilGq46vQePFKoehADWWsDvDCZSNsMU7PrpPr3EM3lU/edit#gid=0

If people can complete this doc and send an email to the dev mailing list then 
we can formalize the features and commit them to a doc in the OVS repo.

It would be good if others are also willing to share their plans so that we can 
build up a complete picture of what's planned for the OVS 2.9 release and make 
sure there's no duplication.

Keepalive: Keepalive feature is aimed at achieving Fastpath Service Assurance  
in OVS-DPDK deployments. It adds support for monitoring the packet processing 
threads by dispatching heartbeats at regular intervals and the status is 
updated periodically in OVSDB. In case of heartbeat misses the failure is 
detected and reported to higher level fault management systems/frameworks. The 
status can be monitored from OpenStack Ceilometer service.

vHost PMD: The vHost PMD brings vHost User ports under control of DPDK's 
librte_ether library and in doing so DPDK's librte_vhost library will no longer 
be directly referenced in OVS. Potential benefits include shared code among 
port types in netdev-dpdk and smoother upgrades between DPDK versions in OVS.

Multi-segment Mbuf: Support for jumbo frames was added to netdev-dpdk in 
v2.6.0. In that implementation, a jumbo frame is stored within a single-segment 
mbuf. That mode will remain the default, but an option will now be added to 
allow a jumbo frame to be stored within a multi-segment mbuf. This reduces the 
requirement for large, contiguous blocks of memory, and may be useful in 
deployments with limited memory.

Conn Track: Analyze and improve Connection Tracker performance:
The Connection Tracker is a feature to manage stateful connections and 
implement security Firewalls.
This allows a better protection against attacks and helps in load balancing. 
The counterpart is a significant impact on the overall performance. This work 
is aimed at analyzing possible bottlenecks - also considering the latest 
protocol implementations - to improve the Connection Tracker performance.

IPSEC: This feature looks to introduce IPsec into OVS with DPDK. IPsec would 
function in transport mode and would be used in conjunction with existing 
encapsulation methods (initially VxLAN) to create a new interface type 
'vxlanipsec'. The DPDK cryptodev framework will be used to  handle 
cipher/digest operations as part of the encap/decap actions in accordance with 
a security association. The cryptodev devices supported would be limited to 
virtual crypto devices such as the  AESN-MB vdev. As such cipher and digest 
algorithms supported would be limited to the capabilities of the vdev. In terms 
of Security Association generation for IPsec policies, the feature will allow a 
user to specify security associations via command line options for the 
interface.

Zero copy: Support for enabling DPDK's 'dequeue zero copy' feature on vHost 
User ports which removes the costly memcpy operation from guest to host when 
enabled. Detailed information in the DPDK documentation: 
http://dpdk.org/doc/guides/prog_guide/vhost_lib.html#vhost-api-overview.

DPDK 17.05.2 and 17.11 support: DPDK 17.05.2 is the latest stable release of 
DPDK, and includes numerous bug-fixes and stability updates.
DPDK 17.11 will be the next DPDK Long Term Support (LTS) release (TBD). It will 
also include numerous new features, such as Generic Segmentation Offload (GSO) 
and vHost PMD.

Upcall performance: The first packet in every flow that is handled by OVS is an 
upcall to the exception path. This is a performance bottleneck for some use 
cases. Analysis and improvements to the performance of this code path will be 
investigated.

Service Assurance : Virtual switching reporting of flow telemetry via IPFix.

Extended NIC Stats: Expose low level driver statistics registers to the user 
through the Extended NIC statistics API in DPDK for physical interfaces.

NAT extended counters (sflow): Monitoring NAT flow telemetry via sFlow.

___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev