|
Re: audience for "reference architecture" content
alexis richardson
On Fri, 22 Jul 2016, 17:12 Alan Conley, <aconley@...> wrote:
yes I do think that network and a couple of other things are, ideally, pluggable in each platform.
|
|
|
|
Re: audience for "reference architecture" content
Alan Conley <aconley@...>
Of course they start with one of the major monolithic solutions in this space, k8s, mesos, docker dc. However, they are working with different networking, monitoring, service r&d solutions. Is there no desire to allow companies to replace a component in one of these major solutions? From: Alexis Richardson <alexis@...>
Sent: Friday, July 22, 2016 9:05 AM To: Alan Conley Cc: cncf-toc@... Subject: Re: audience for "reference architecture" content On Fri, Jul 22, 2016 at 5:01 PM, Alan Conley <aconley@...> wrote:
> I'm referring mostly to large enterprises. I find that startling. Are you sure they are not just creating integration points around the 'edge' of Kubernetes or Docker Data Center or ... > > > Alan > > > > ________________________________ > From: Alexis Richardson <alexis@...> > Sent: Friday, July 22, 2016 8:50 AM > To: Alan Conley > Cc: cncf-toc@... > Subject: Re: audience for "reference architecture" content > > Alan, > > Thank-you. > > "anyone assembling a container management solution (orchestrator, > control plane etc.) and anyone attempting to build a functional > component that would fit within that architecture. Pretty much every > company I've spoken with is currently rolling their own from a select > set of open source projects and developing their own glue to fill gaps > and stitch the pieces together" > > Does this mean: > - large enterprises > - vendors > - other..? > > a > > > > On Fri, Jul 22, 2016 at 4:44 PM, Alan Conley <aconley@...> wrote: >> I think this goes back to what is the charter of the CNCF. >> >> >> Taking a few short cuts in this narrative and avoiding some of the >> politics. >> Containers became mainstream when docker "standardized" their use. The >> industry saw this rapid adoption and suggested a forum should "govern" >> these >> standards which resulted in the OCP->OCI. From the OCI charter, >> "...industry participants may easily contribute to building a >> vendor-neutral, portable and open specification and runtime...". So we >> have >> both a spec and working code. >> >> >> What was missing, was the equivalent for container management >> (orchestrator, >> control plane, monitoring) solutions. I believe this was the genesis of >> the >> CNCF, originated by Craig and why k8s was the initial project. The >> original >> reference architecture provided a simple view of these functional >> components. BTW, most would see the similarities between this and >> OpenStack >> for VMs. I personally have no interest in seeing the CNCF focused on one >> implementation. >> >> >> Assuming I'm not completely off the path, the target audience for the ref >> arch is anyone assembling a container management solution (orchestrator, >> control plane etc.) and anyone attempting to build a functional component >> that would fit within that architecture. Pretty much every company I've >> spoken with is currently rolling their own from a select set of open >> source >> projects and developing their own glue to fill gaps and stitch the pieces >> together. (My company is doing that for our own SaaS solutions.) There >> are >> a few of us interested in extending what we see as one of the needed >> functional components. However, we are unclear on how others see that >> interfacing with other components and in some cases see overlapping >> capabilities. We can certainly just put it out there and see if there is >> adoption, but then that begs the question as to what value does the CNCF >> actually provide other than marketing. >> >> >> I'll stop before this becomes too much of a ramble, for comments. >> >> >> Alan >> >> >> >> >> >> ________________________________ >> From: Alexis Richardson <alexis@...> >> Sent: Thursday, July 21, 2016 4:54 AM >> To: cncf-toc@... >> Subject: audience for "reference architecture" content >> >> Hi all, >> >> Yesterday we had our 2nd discussion about the 'marketecture' stack that >> Ken >> and I put out. One piece of feedback, from Doug Davis by email, and then >> eg >> from Alan Conley on the call, was that much more detail could help. >> >> I believe this is a "target audience" issue. We may need different >> material >> for different audiences even if those audiences are all "technical". For >> example - I argued that for developer end users, less detail is good. >> >> It would be very helpful to hear from the people advocating for more >> detail, >> regarding their target audience. >> >> alexis >> >> >> |
|
|
|
Re: audience for "reference architecture" content
alexis richardson
On Fri, Jul 22, 2016 at 5:01 PM, Alan Conley <aconley@...> wrote:
I'm referring mostly to large enterprises.I find that startling. Are you sure they are not just creating integration points around the 'edge' of Kubernetes or Docker Data Center or ...
|
|
|
|
Re: audience for "reference architecture" content
Alan Conley <aconley@...>
I'm referring mostly to large enterprises.
Alan From: Alexis Richardson <alexis@...>
Sent: Friday, July 22, 2016 8:50 AM To: Alan Conley Cc: cncf-toc@... Subject: Re: audience for "reference architecture" content Alan,
Thank-you. "anyone assembling a container management solution (orchestrator, control plane etc.) and anyone attempting to build a functional component that would fit within that architecture. Pretty much every company I've spoken with is currently rolling their own from a select set of open source projects and developing their own glue to fill gaps and stitch the pieces together" Does this mean: - large enterprises - vendors - other..? a On Fri, Jul 22, 2016 at 4:44 PM, Alan Conley <aconley@...> wrote: > I think this goes back to what is the charter of the CNCF. > > > Taking a few short cuts in this narrative and avoiding some of the politics. > Containers became mainstream when docker "standardized" their use. The > industry saw this rapid adoption and suggested a forum should "govern" these > standards which resulted in the OCP->OCI. From the OCI charter, > "...industry participants may easily contribute to building a > vendor-neutral, portable and open specification and runtime...". So we have > both a spec and working code. > > > What was missing, was the equivalent for container management (orchestrator, > control plane, monitoring) solutions. I believe this was the genesis of the > CNCF, originated by Craig and why k8s was the initial project. The original > reference architecture provided a simple view of these functional > components. BTW, most would see the similarities between this and OpenStack > for VMs. I personally have no interest in seeing the CNCF focused on one > implementation. > > > Assuming I'm not completely off the path, the target audience for the ref > arch is anyone assembling a container management solution (orchestrator, > control plane etc.) and anyone attempting to build a functional component > that would fit within that architecture. Pretty much every company I've > spoken with is currently rolling their own from a select set of open source > projects and developing their own glue to fill gaps and stitch the pieces > together. (My company is doing that for our own SaaS solutions.) There are > a few of us interested in extending what we see as one of the needed > functional components. However, we are unclear on how others see that > interfacing with other components and in some cases see overlapping > capabilities. We can certainly just put it out there and see if there is > adoption, but then that begs the question as to what value does the CNCF > actually provide other than marketing. > > > I'll stop before this becomes too much of a ramble, for comments. > > > Alan > > > > > > ________________________________ > From: Alexis Richardson <alexis@...> > Sent: Thursday, July 21, 2016 4:54 AM > To: cncf-toc@... > Subject: audience for "reference architecture" content > > Hi all, > > Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken > and I put out. One piece of feedback, from Doug Davis by email, and then eg > from Alan Conley on the call, was that much more detail could help. > > I believe this is a "target audience" issue. We may need different material > for different audiences even if those audiences are all "technical". For > example - I argued that for developer end users, less detail is good. > > It would be very helpful to hear from the people advocating for more detail, > regarding their target audience. > > alexis > > > |
|
|
|
Re: audience for "reference architecture" content
alexis richardson
Alan,
toggle quoted message
Show quoted text
Thank-you. "anyone assembling a container management solution (orchestrator, control plane etc.) and anyone attempting to build a functional component that would fit within that architecture. Pretty much every company I've spoken with is currently rolling their own from a select set of open source projects and developing their own glue to fill gaps and stitch the pieces together" Does this mean: - large enterprises - vendors - other..? a On Fri, Jul 22, 2016 at 4:44 PM, Alan Conley <aconley@...> wrote:
I think this goes back to what is the charter of the CNCF. |
|
|
|
Re: audience for "reference architecture" content
Alan Conley <aconley@...>
I think this goes back to what is the charter of the CNCF.
Taking a few short cuts in this narrative and avoiding some of the politics. Containers became mainstream when docker "standardized" their use. The industry saw this rapid adoption and suggested a forum should "govern" these standards which resulted in the OCP->OCI. From the OCI charter, "...industry participants may easily contribute to building a vendor-neutral, portable and open specification and runtime...". So we have both a spec and working code.
What was missing, was the equivalent for container management (orchestrator, control plane, monitoring) solutions. I believe this was the genesis of the CNCF, originated by Craig and why k8s was the initial project. The original reference architecture provided a simple view of these functional components. BTW, most would see the similarities between this and OpenStack for VMs. I personally have no interest in seeing the CNCF focused on one implementation.
Assuming I'm not completely off the path, the target audience for the ref arch is anyone assembling a container management solution (orchestrator, control plane etc.) and anyone attempting to build a functional component that would fit within that architecture. Pretty much every company I've spoken with is currently rolling their own from a select set of open source projects and developing their own glue to fill gaps and stitch the pieces together. (My company is doing that for our own SaaS solutions.) There are a few of us interested in extending what we see as one of the needed functional components. However, we are unclear on how others see that interfacing with other components and in some cases see overlapping capabilities. We can certainly just put it out there and see if there is adoption, but then that begs the question as to what value does the CNCF actually provide other than marketing.
I'll stop before this becomes too much of a ramble, for comments.
Alan
From: Alexis Richardson <alexis@...>
Sent: Thursday, July 21, 2016 4:54 AM To: cncf-toc@... Subject: audience for "reference architecture" content Hi all,
Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken and I put out. One piece of feedback, from Doug Davis by email, and then eg from Alan Conley on the call, was that much more detail could help.
I believe this is a "target audience" issue. We may need different material for different audiences even if those audiences are all "technical". For example - I argued that for developer end users, less detail is good.
It would be very helpful to hear from the people advocating for more detail, regarding their target audience.
alexis
|
|
|
|
Re: audience for "reference architecture" content
alexis richardson
All, Here is a slightly cleaned up version of the "projects and products" data that I mentioned. It is also a cloud native market landscape. Ken and I used this to inform our initial thinking on the simplified reference stack, which is also shown on the second tab. Market landscape sheet: https://docs.google.com/spreadsheets/d/1ify0vCXxum_TKtA99neDWhciZeUbnp8dSNqVGXsQEeA/edit#gid=1990705469 The original categories were based on public analysis by Matt Miller at Sequoia, but I have modified the categorisation a little. I have also shown this data to other folks who analyse the space - press, investors, etc. It feels reasonably thorough. I ask that you be gracious in your comments, first by noting some caveats. 1) Some of the categories are definitely "not quite right". For example the APM-Logging spectrum is a moving target. I have tried to split this into functions like "visualization" and "monitoring", without complete success. It feels unfair in places. Same thing for "platform". But, I think it's a starting point. 2) I have not tried to put projects in "order of significance". Just because something is at the top of a list, it does not make it more important than stuff lower down. The ordering is more "as I thought of them" - ie. subjective. It would be good to have a better model here, eg. public vs private vs startup vs community... But even that could be invidious. 3) The data in the sheet does not *quite* line up yet with some of the items in Ken's v0.3 deck. https://docs.google.com/presentation/d/1uMw2wkK0ubmc3khxqIuxK_rLK_wN89tNCnK7gDmTGR8/edit#slide=id.p8 alexis PS -- Doug, I hope this helps with your questions below. On Thu, Jul 21, 2016 at 1:28 PM, Doug Davis <dug@...> wrote:
|
|
|
|
Re: audience for "reference architecture" content
Doug Davis <dug@...>
Alexis, Hi all, Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken and I put out. One piece of feedback, from Doug Davis by email, and then eg from Alan Conley on the call, was that much more detail could help. I believe this is a "target audience" issue. We may need different material for different audiences even if those audiences are all "technical". For example - I argued that for developer end users, less detail is good. It would be very helpful to hear from the people advocating for more detail, regarding their target audience. alexis _______________________________________________ cncf-toc mailing list cncf-toc@... https://lists.cncf.io/mailman/listinfo/cncf-toc |
|
|
|
audience for "reference architecture" content
alexis richardson
Hi all, Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken and I put out. One piece of feedback, from Doug Davis by email, and then eg from Alan Conley on the call, was that much more detail could help. I believe this is a "target audience" issue. We may need different material for different audiences even if those audiences are all "technical". For example - I argued that for developer end users, less detail is good. It would be very helpful to hear from the people advocating for more detail, regarding their target audience. alexis |
|
|
|
introducing - anand from minio
alexis richardson
Hi all, There was some interest in talking with people who work on Minio. I have copied Anand "AB" Periasamy, with whom I spoke on Tuesday. AB, On the TOC call yesterday, I introduced the idea of Minio creating a CNCF Proposal. I want to give you some early feedback on this. Please note that TOC calls are public and that this CNCF mailing list is public. The first impression - as I mentioned when we spoke on Tuesday - is that Minio is a 'young' project. As a TOC we are probably minded towards more mature projects at present. One question that arose is - who is using Minio in production? There was a feeling that this is especially a concern with distributed storage technology where maturity is hard won. Overall, cloud native storage is quite interesting. We'd love to know: what new things Minio is doing that make it special, how it is designed, and so on. Do you have anything you could share on that topic? alexis |
|
|
|
when is my app "cloud native"
alexis richardson
hi all, there was a fun convo on slack yesterday, that I thought it might be of interest to the mailing list.... a Abstract question: if one were to 'check' a source project for cloud-nativeness (in some automatable way) - where would they start? Alexis good question! various answers spring to mind. can it be run on any cloud and virtual data center does it support dynamic / elastic use does it use containers and microservices and programmable infra can deployment and management be automated at any scale can it be upgraded and downgraded without losing availability does the data share the application's fate or is it resilient Kit Cool. Yep, those are great 'checks'. Now thinking about what to look for in a source repo (say during a CI/CD process) that would indicate cloud-nativeness. RE: containers - could just look for the presence of a Dockerfile (or similar) but, some of these things are outside the source/project - and might be more CM-oriented. Alexis I wouldn't take much from having a Dockerfile Kit Of course, I was just using that as an example. I'm thinking it has to be passive - can't use some definition file that could just be 'whipped'. I think what I am coming to is that there isn't a way to just look at a source repo - that there's a bigger picture, likely organizational that is more indicative. |
|
|
|
Re: next steps on "opening up the cluster"
Zachary Smith
All,
Sorry I wasn't on the call today, traveling in Japan now (sorry I missed you here Chris!). I think an issue to highlight here is that we're running into very typical service provider things around effectuating access, managing capacity, good accounting/oversight, etc. Most of these issues (especially with a wider audience) creep up after the initial access and could become a burden for the CNCF or create a negative experience for the users if not handled properly. Given what we do at Packet, I'm intimate with some of the challenges here. A few questions I have: -Does the CNCF have the ability and resources to act as a service provider to its members for this? Grant access, reset access, respond to questions of a wider audience and their users, etc? -Is self service an important part of the access that end user projects (and their users, developers, etc) need in order to make successful use of the resources? -Can Packet throw its hat in the ring as a consumer of some of the resources that we would then offer out via our platform to users / organizations that CNCF approves? At Packet we have several dozen related projects (some CNCF members, some not) that we sponsor with free on-demand bare metal infrastructure for similar goals as the CNCF, e.g. to perform scale testing, setup non VM-based workloads, to test raw networking and related stacks, etc. So my selfish ask would be that some of this could actually end up on the CNCF cluster vs our own capacity, but only if that is deemed useful to the CNCF and it's membership goals. The non-selfish ask is that we'd love to add some value around this granting of access, provisioning and ongoing accounting/resource management if that is of use. I'd love to have a further chat on this subject if there is interest. thanks! -Zac
--
|
|
|
|
Re: DRAFT slides for TOC meeting tomorrow
NASSAUR, DOUGLAS C <dn283x@...>
I'm almost done with draft one of the periodic table of cloud native elements for you guys to throw rocks at. We should then align with ref arch
Regards, Doug
|
|
|
|
Re: next steps on "opening up the cluster"
Donaldson, Jonathan <jonathan.donaldson@...>
To be clear. The cluster is open to all CNCF members. Submit the request as listed below, and the governance committee will respond and facilitate.
Thank you,
Jonathan Donaldson VP/GM, Software Defined Infrastructure Group
|
|
|
|
Re: next steps on "opening up the cluster"
Jason Mendenhall
I think it's important to ensure that we don't end up with projects scattered across multiple locations.
The objective should always be to put all projects on the cluster. In my opinion, it's one of our strengths as an organization.
We currently have a multi-tenant cloud environment for provisioning VMs and a section where dedicated hardware can be leveraged.
We have contemplated the idea that we should provide $0/timed access to the multi-tenant space "the CNCF cloud" to all members. In essence, you get a login and off you go and all dedicated hardware projects would follow the process
Chris outlined.
Are we deviating from that approach? Our we missing some execution steps to opening it up? Does the cluster committee need to assemble to discuss how we might do this?
I've been out for a couple of weeks so I may be out of the loop. If any of this is redundant, apologies from the clueless guy.
Jason M.
702.333.6570
CONFIDENTIAL INFORMATIONThis email message, its chain, and any attachments: (a) may include proprietary information, trade secrets, confidential information and/or other protected information ("Confidential Information") which are hereby labeled as Confidential for protection purposes, (b) is sent to you in confidence with a reasonable expectation of privacy, (c) may be protected by confidentiality agreements requiring this notice and/or identification, and (d) is not intended for transmission to, or receipt by unauthorized persons. If you are not the intended recipient, please notify the sender immediately by telephone or by replying to this message. Please then delete this message, any attachments, chains, copies or portions from your system(s). Thank you. CONFIDENTIAL INFORMATIONThis email message, its chain, and any attachments: (a) may include proprietary information, trade secrets, confidential information and/or other protected information ("Confidential Information") which are hereby labeled as Confidential for protection purposes, (b) is sent to you in confidence with a reasonable expectation of privacy, (c) may be protected by confidentiality agreements requiring this notice and/or identification, and (d) is not intended for transmission to, or receipt by unauthorized persons. If you are not the intended recipient, please notify the sender immediately by telephone or by replying to this message. Please then delete this message, any attachments, chains, copies or portions from your system(s). Thank you. |
|
|
|
Re: next steps on "opening up the cluster"
NASSAUR, DOUGLAS C <dn283x@...>
I've proposed a "reference architecture" environment people could experience by "renting" time (at $0) to illustrate cloud native vs traditional virtualization approaches like openstack. Several of us have created and would donate the illustrative solution.
Regards, Doug
|
|
|
|
Re: next steps on "opening up the cluster"
i sense a CNCF Cloud in the works ;-) Diane On Wed, Jul 20, 2016 at 9:24 AM, Solomon Hykes <solomon.hykes@...> wrote: This would be very useful for the Docker developers as well! They would love to have access too, would that be ok? --
Kind Regards, Diane Mueller Director, Community Development Red Hat OpenShift @openshiftcommons We have more in Common than you know, learn more at http://commons.openshift.org |
|
|
|
Re: next steps on "opening up the cluster"
Solomon Hykes <solomon.hykes@...>
This would be very useful for the Docker developers as well! They would love to have access too, would that be ok?
toggle quoted message
Show quoted text
On Wednesday, July 20, 2016, Alexis Richardson via cncf-toc <cncf-toc@...> wrote:
|
|
|
|
Re: next steps on "opening up the cluster"
alexis richardson
yes it could On Wed, 20 Jul 2016, 17:04 Diane Mueller-Klingspor, <dmueller@...> wrote:
|
|
|
|
Re: next steps on "opening up the cluster"
All, The Fedora team has been asking for access to the cncf cluster for some testing of their Fedora cloud projects. Could this "opening" up include them as a potential guest on the cluster? Diane Mueller Director, Community Development Red Hat OpenShift @openshiftcommon (604) 765 3635 On Wed, Jul 20, 2016 at 8:54 AM, Chris Aniszczyk via cncf-toc <cncf-toc@...> wrote:
--
Kind Regards, Diane Mueller Director, Community Development Red Hat OpenShift @openshiftcommons We have more in Common than you know, learn more at http://commons.openshift.org |
|
|
Alexis Richardson via cncf-toc ---07/21/2016 07:54:53 AM---Hi all, Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken
Alexis Richardson via cncf-toc ---07/21/2016 07:54:53 AM---Hi all, Yesterday we had our 2nd discussion about the 'marketecture' stack that Ken