You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by Axel Baudot <ax...@protonmail.com.INVALID> on 2023/04/20 23:13:35 UTC
Arguing for Cloudstack for a HPC oriented datacenter
Hello dear CloudStack users,
As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
In particular I have the following concerns :
1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed?
2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination.
3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard?
4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case?
Thanks for reading, any food for thoughts will be very much appreciated.
Best,
Axel
Re: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Nux <nu...@li.nux.ro>.
Hello,
My Cloudstack journey as a sysadmin and (ex)cloud operator started with
Openstack. I've since been through OpenNebula, oVirt etc.
I'm obviously wearing the Cloudstack hat now, but that's because it hits
the sweet spot between features and complexity, especially if you're
with a small ops team.
I don't see anything in the general HPC requirements that Cloudstack
cannot satisfy, same for your points below.
You'll need:
- Scalability - checked
- Flexibility - checked
- Cost effectiveness - dooh
- Resource management - checked
- Multi-tenancy - checked
- Interoperability and automation - checked (open, fully documented api,
automation tools available - terraform, ansible etc)
- Hearty community support - checked
- Commercial support, should you need it - checked
As far as k8s is concerned, I think CAPI/CAPC can be an answer.
On Kubeflow, should you wish to use it - knowing nothing about it other
that it's ML focused - you will need to see what are the GPU
requirements there, if any, that could be a pain point, albeit surely
addressable one way or another.
Taking my Cloudstack hat off, in an ideal world you would take some time
to test the 3 solutions thoroughly (cloudstack, openstack, opennebula)
and see which matches your needs and capabilities best and also have a
go at maintainability, because getting off the ground is the easy bit.
And lastly, if you have paid attention to this project, you will have
noticed that it's a living, breathing one, if something is wrong or
missing today, it will be fixed tomorrow, metaphorically speaking.
Looking at the roadmap of the past few years, it's just crazy how many
new awesome things materialised.
HTH
On 2023-04-21 00:13, Axel Baudot wrote:
> Hello dear CloudStack users,
>
> As a team sitting at a crossroad to chose a solution to provision
> ressources for a HPC oriented datacenter: can a strong case be made for
> the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
>
> In particular I have the following concerns :
>
> 1. The company provides infrastructure to clients, as well as full
> blown project development, so it trying to set up both a public cloud
> with resource billing and an internal development platform. This is the
> use case covered by CloudStack, right? Or could the flexibility of
> OpenStack be needed?
>
> 2. It is also heavily geared toward HPC. I have seen little mention of
> HPC with CloudStack, while it is heavily advertised in OpenStack world,
> due to the ties with CERN and such. Can you think of limitations of
> CloudStack for this use case? I am also interested in case studies or
> any reading material on this combination.
>
> 3. The team could standardize on containers soon (especially for dev
> environments). I would tend to do a simple VM + Docker workflow. Is it
> standard?
>
> 4. I am all for simplicity of operation and maintenance. I think
> CloudStack could really shine here, right? I also think that Kubernetes
> should be avoided because of this. Thing is, Kubeflow is getting some
> attention and we might have to support it... Is CKS considered mature
> for production and a viable solution in this case?
>
> Thanks for reading, any food for thoughts will be very much
> appreciated.
>
> Best,
> Axel
RE: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Alex Mattioli <Al...@shapeblue.com>.
Adding my 2cents. Inline with your questions.
-----Original Message-----
From: Axel Baudot <ax...@protonmail.com.INVALID>
Sent: Friday, April 21, 2023 1:14 AM
To: users <us...@cloudstack.apache.org>
Subject: Arguing for Cloudstack for a HPC oriented datacenter
Hello dear CloudStack users,
As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
In particular I have the following concerns :
1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed?
>>Cloudstack is absolutely fit for purpose there, I've used it myself for exactly that, HPC and HPTC.
2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination.
>HPC has two sides:
1) a collapsed core backbone network, 100Gb+ interfaces, extremely low latency switches, NVMe based storage arrays with Fibre Channel connectivity, etc...etc... CloudStack is completely neutral to that and adds no performance impact in these domains.
2) GPU processing - CloudStack has support for GPU passthrough and new GPU models can be added relatively easily.
Distributed network processing - CloudStack now supports Tungsten Fabric, so you can have routed traffic in extreme scales without bottlenecks. Or, simply use pure L2 networks.
Specialized templates - CloudStack doesn't limit that
High CPU count - CloudStack doesn't limit that, it's only a hypervisor limit
3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard?
>Absolutely standard
4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case?
Thanks for reading, any food for thoughts will be very much appreciated.
Best,
Axel
RE: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Alex Mattioli <Al...@shapeblue.com>.
>> CloudStack is considered to be more lightweight and easier to deploy and manage, while OpenStack is more flexible and >>customizable.
I fully agree with the first part of the sentence, but the second isn't really true, CloudStack is as customizable and flexible as OpenStack. It is less modular than OpenStack, but has a very robust plugin framework that more than make up for that.
The one thing that OpenStack definitely has that CloudStack doesn't is the marketing budget, and there's where you get all kinds of use cases and whitepapers coming from.
I'd be really surprised if anyone had any serious and non-niche use case where OpenStack is actually a better fit than CloudStack.
Cheers
Alex
-----Original Message-----
From: Axel Baudot <ax...@protonmail.com.INVALID>
Sent: Friday, April 21, 2023 11:11 PM
To: users <us...@cloudstack.apache.org>
Subject: Re: Arguing for Cloudstack for a HPC oriented datacenter
Thanks to everyone who chimed in, those are all great answers and valuable info.
Giles, my comment about flexibility stems mostly from having no practical experience with CloudStack while I have worked with a few incarnations of OpenStack that differed quite a lot in features. Also, Hongtu Zang in a previous thread made the following comment:
> CloudStack is considered to be more lightweight and easier to deploy and manage, while OpenStack is more flexible and customizable.
It seemed reasonable that the two aspects could be at odd and a trade-off would have to be made. Glad to read about multi-tennancy and that architecture flexibility is retained. Best, Axel
------- Original Message -------
On Friday, April 21st, 2023 at 4:29 PM, Giles Sirett wrote: > Hi Axel, I'll try to give my view on your questions > > Kind Regards > Giles > > > > > -----Original Message----- > From: Axel Baudot axel.baudot@protonmail.com.INVALID > > Sent: Friday, April 21, 2023 1:14 AM > To: users users@cloudstack.apache.org > > Subject: Arguing for Cloudstack for a HPC oriented datacenter > > Hello dear CloudStack users, > > As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes? > > In particular I have the following concerns : > > 1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed? > > > > This is well within Cloudstacks scope. As Cloudstack is multi-tennant from the ground-up, the internal dev platform would likely be just another tennant. It may be implemented as different zones, different domains or a whole range of different approaches but they are simply architectural choices > > > > I'd be interested to ask what extra flexibility you'd think you'd get with open stack for this use case ? > > > 2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination. > > > > From a storage and compute perspective, that is really a conversation about the hardware and the hypervisor, not about what you use to orchestrate it. Networking is really the thing to consider with HPC and orchestration. Cloudstacks VR can scale but does need to be considered. > > > > 3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard? > > > > Hmmm - very hard to answer that without knowing more. Most orgs these days would use K8S to orchestrate at the container level, with Cloudstack underneath (using CloudStacks CAPI provider or CKS) > > > 4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case? > > > > Absolutely. Lots of orgs using CKS in production, but bear in mind that CKS is just deploying K8S clusters - so it doesn't avoid Kubernetes - it just makes deploying K8S simple > > > Thanks for reading, any food for thoughts will be very much appreciated. > > > > More generally, if you want to compare Openstack, Cloudstack and opennebula - you should really do a side-by-side evaluation > > > The reason people choose CloudStack over openstack is usually ease/cost/hassle - nearly everybody has to deploy a commercial distribution (or have a massive team). Cloudstack is "pure" opensource - everybody running CloudStack is running >>the opensource code and you can usually deploy Cloudstack in production in the same time as it takes to POC openstack . Opennebula isn’t as complicated but I don’t know if anybody is running it without the commercial version > > > > > > > Best, > Axel
Re: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Axel Baudot <ax...@protonmail.com.INVALID>.
Thanks to everyone who chimed in, those are all great answers and valuable info.
Giles, my comment about flexibility stems mostly from having no practical experience with CloudStack while I have worked with a few incarnations of OpenStack that differed quite a lot in features. Also, Hongtu Zang in a previous thread made the following comment:
> CloudStack is considered to be more lightweight and easier to deploy and manage, while OpenStack is more flexible and customizable.
It seemed reasonable that the two aspects could be at odd and a trade-off would have to be made. Glad to read about multi-tennancy and that architecture flexibility is retained. Best, Axel
------- Original Message -------
On Friday, April 21st, 2023 at 4:29 PM, Giles Sirett wrote: > Hi Axel, I'll try to give my view on your questions > > Kind Regards > Giles > > > > > -----Original Message----- > From: Axel Baudot axel.baudot@protonmail.com.INVALID > > Sent: Friday, April 21, 2023 1:14 AM > To: users users@cloudstack.apache.org > > Subject: Arguing for Cloudstack for a HPC oriented datacenter > > Hello dear CloudStack users, > > As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes? > > In particular I have the following concerns : > > 1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed? > > > > This is well within Cloudstacks scope. As Cloudstack is multi-tennant from the ground-up, the internal dev platform would likely be just another tennant. It may be implemented as different zones, different domains or a whole range of different approaches but they are simply architectural choices > > > > I'd be interested to ask what extra flexibility you'd think you'd get with open stack for this use case ? > > > 2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination. > > > > From a storage and compute perspective, that is really a conversation about the hardware and the hypervisor, not about what you use to orchestrate it. Networking is really the thing to consider with HPC and orchestration. Cloudstacks VR can scale but does need to be considered. > > > > 3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard? > > > > Hmmm - very hard to answer that without knowing more. Most orgs these days would use K8S to orchestrate at the container level, with Cloudstack underneath (using CloudStacks CAPI provider or CKS) > > > 4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case? > > > > Absolutely. Lots of orgs using CKS in production, but bear in mind that CKS is just deploying K8S clusters - so it doesn't avoid Kubernetes - it just makes deploying K8S simple > > > Thanks for reading, any food for thoughts will be very much appreciated. > > > > More generally, if you want to compare Openstack, Cloudstack and opennebula - you should really do a side-by-side evaluation > > > The reason people choose CloudStack over openstack is usually ease/cost/hassle - nearly everybody has to deploy a commercial distribution (or have a massive team). Cloudstack is "pure" opensource - everybody running CloudStack is running >>the opensource code and you can usually deploy Cloudstack in production in the same time as it takes to POC openstack . Opennebula isn’t as complicated but I don’t know if anybody is running it without the commercial version > > > > > > > Best, > Axel
RE: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Giles Sirett <gi...@shapeblue.com>.
Hi Axel, I'll try to give my view on your questions
Kind Regards
Giles
-----Original Message-----
From: Axel Baudot <ax...@protonmail.com.INVALID>
Sent: Friday, April 21, 2023 1:14 AM
To: users <us...@cloudstack.apache.org>
Subject: Arguing for Cloudstack for a HPC oriented datacenter
Hello dear CloudStack users,
As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
In particular I have the following concerns :
1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed?
>> This is well within Cloudstacks scope. As Cloudstack is multi-tennant from the ground-up, the internal dev platform would likely be just another tennant. It *may* be implemented as different zones, different domains or a whole range of different approaches but they are simply architectural choices
>> I'd be interested to ask what extra flexibility you'd think you'd get with open stack for this use case ?
2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination.
>>From a storage and compute perspective, that is really a conversation about the hardware and the hypervisor, not about what you use to orchestrate it. Networking is really the thing to consider with HPC and orchestration. Cloudstacks VR can scale but does need to be considered.
3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard?
>>Hmmm - very hard to answer that without knowing more. Most orgs these days would use K8S to orchestrate at the container level, with Cloudstack underneath (using CloudStacks CAPI provider or CKS)
4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case?
>>Absolutely. Lots of orgs using CKS in production, but bear in mind that CKS is just deploying K8S clusters - so it doesn't avoid Kubernetes - it just makes deploying K8S simple
Thanks for reading, any food for thoughts will be very much appreciated.
>> More generally, if you want to compare Openstack, Cloudstack and opennebula - you should really do a side-by-side evaluation
>>The reason people choose CloudStack over openstack is usually ease/cost/hassle - nearly everybody has to deploy a commercial distribution (or have a massive team). Cloudstack is "pure" opensource - everybody running CloudStack is running >>the opensource code and you can usually deploy Cloudstack in production in the same time as it takes to POC openstack . Opennebula isn’t as complicated but I don’t know if anybody is running it without the commercial version
Best,
Axel
Re: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Axel Baudot <ax...@protonmail.com.INVALID>.
Hello Rohit,
Thank you for your detailed answer and many references. I am excited to take Cloudstack for a test drive as per your instructions.
Axel
-------- Original Message --------
On Apr 21, 2023, 6:51 AM, Rohit Yadav wrote:
> In addition, Axel, you can also look at: https://cloudstack.apache.org/cloud-builders.html https://cloudstack.apache.org/kubernetes.html Regards. ________________________________ From: Rohit Yadav Sent: Friday, April 21, 2023 16:16 To: users Subject: Re: Arguing for Cloudstack for a HPC oriented datacenter Hi Alex, Please find my answers in-line: > As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes? Yes, absolutely! > 1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed? There are many public and private cloud service providers, enterprises, and organisations who use CloudStack as their IaaS cloud platform. You can see the known-users list here: https://cloudstack.apache.org/users.html (we're opensource, and we don't track who uses CloudStack, unless somebody volunteers to put the users name on the website). In terms of resource/billing portals, there quite a few vendors I'm sure you can search them and see some demos/presentations for example https://www.youtube.com/results?search_query=cloudstack+billing > 2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination. Yes, some of the largest clouds with HPC and similar requirements are using CloudStack. You may find several use cases here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Case+Studies and talks on youtube such as https://www.youtube.com/watch?v=xq2LVY18GU8 and talks from recent CCC https://www.youtube.com/playlist?list=PLnIKk7GjgFlYcKNbhYSgWGbJ8nJ0g496V > 3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard? Yes, this can be used. CloudStack IaaS platform doesn't care what you run inside the VM - you're free to decide on your stack and development choices. > 4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case? CloudStack is boring and just works. Of course, I encourage you to setup a test env. and play with it, test and try all the requirements and see for yourself. On container orchestration, you can setup and run whatever you'd like in VMs deployed via CloudStack or you can also use (a) CKS which is CloudStack's managed Kubernetes service offering, but we now also have (b) CAPC https://cluster-api-cloudstack.sigs.k8s.io. CAPC even works with EKS-A you can try. Both options are considered production and viable solution, but each have their pros and cons. If you're looking for more flexbility and not get tied into any IaaS platform then CAPC is the way to go. My suggestion is for you to test drive CloudStack. You can try: https://github.com/shapeblue/hackerbook/blob/main/1-user.md You can follow our quickstart guide here: https://docs.cloudstack.apache.org/en/latest/quickinstallationguide/qig.html or if you want an Ubuntu-based guide here; https://rohityadav.cloud/blog/cloudstack-kvm/ If you've an Ubuntu desktop/laptop, you can even try mbx https://github.com/shapeblue/mbx to deploy CloudStack with one of KVM/XenServer/XCP-ng/VMware hypervisors (in a nested env, i.e. on a VM) with a community or 3rd party repo (see https://cloudstack.apache.org/downloads.html). Regards.
Re: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Rohit Yadav <ro...@shapeblue.com>.
In addition, Axel, you can also look at:
https://cloudstack.apache.org/cloud-builders.html
https://cloudstack.apache.org/kubernetes.html
Regards.
________________________________
From: Rohit Yadav <ro...@shapeblue.com>
Sent: Friday, April 21, 2023 16:16
To: users <us...@cloudstack.apache.org>
Subject: Re: Arguing for Cloudstack for a HPC oriented datacenter
Hi Alex,
Please find my answers in-line:
> As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
Yes, absolutely!
> 1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed?
There are many public and private cloud service providers, enterprises, and organisations who use CloudStack as their IaaS cloud platform. You can see the known-users list here: https://cloudstack.apache.org/users.html (we're opensource, and we don't track who uses CloudStack, unless somebody volunteers to put the users name on the website).
In terms of resource/billing portals, there quite a few vendors I'm sure you can search them and see some demos/presentations for example https://www.youtube.com/results?search_query=cloudstack+billing
> 2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination.
Yes, some of the largest clouds with HPC and similar requirements are using CloudStack. You may find several use cases here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Case+Studies and talks on youtube such as https://www.youtube.com/watch?v=xq2LVY18GU8
and talks from recent CCC https://www.youtube.com/playlist?list=PLnIKk7GjgFlYcKNbhYSgWGbJ8nJ0g496V
> 3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard?
Yes, this can be used. CloudStack IaaS platform doesn't care what you run inside the VM - you're free to decide on your stack and development choices.
> 4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case?
CloudStack is boring and just works. Of course, I encourage you to setup a test env. and play with it, test and try all the requirements and see for yourself.
On container orchestration, you can setup and run whatever you'd like in VMs deployed via CloudStack or you can also use (a) CKS which is CloudStack's managed Kubernetes service offering, but we now also have (b) CAPC https://cluster-api-cloudstack.sigs.k8s.io<https://cluster-api-cloudstack.sigs.k8s.io/>. CAPC even works with EKS-A you can try. Both options are considered production and viable solution, but each have their pros and cons. If you're looking for more flexbility and not get tied into any IaaS platform then CAPC is the way to go. My suggestion is for you to test drive CloudStack.
You can try:
https://github.com/shapeblue/hackerbook/blob/main/1-user.md
You can follow our quickstart guide here: https://docs.cloudstack.apache.org/en/latest/quickinstallationguide/qig.html or if you want an Ubuntu-based guide here; https://rohityadav.cloud/blog/cloudstack-kvm/
If you've an Ubuntu desktop/laptop, you can even try mbx https://github.com/shapeblue/mbx to deploy CloudStack with one of KVM/XenServer/XCP-ng/VMware hypervisors (in a nested env, i.e. on a VM) with a community or 3rd party repo (see https://cloudstack.apache.org/downloads.html).
Regards.
Re: Arguing for Cloudstack for a HPC oriented datacenter
Posted by Rohit Yadav <ro...@shapeblue.com>.
Hi Alex,
Please find my answers in-line:
> As a team sitting at a crossroad to chose a solution to provision ressources for a HPC oriented datacenter: can a strong case be made for the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
Yes, absolutely!
> 1. The company provides infrastructure to clients, as well as full blown project development, so it trying to set up both a public cloud with resource billing and an internal development platform. This is the use case covered by CloudStack, right? Or could the flexibility of OpenStack be needed?
There are many public and private cloud service providers, enterprises, and organisations who use CloudStack as their IaaS cloud platform. You can see the known-users list here: https://cloudstack.apache.org/users.html (we're opensource, and we don't track who uses CloudStack, unless somebody volunteers to put the users name on the website).
In terms of resource/billing portals, there quite a few vendors I'm sure you can search them and see some demos/presentations for example https://www.youtube.com/results?search_query=cloudstack+billing
> 2. It is also heavily geared toward HPC. I have seen little mention of HPC with CloudStack, while it is heavily advertised in OpenStack world, due to the ties with CERN and such. Can you think of limitations of CloudStack for this use case? I am also interested in case studies or any reading material on this combination.
Yes, some of the largest clouds with HPC and similar requirements are using CloudStack. You may find several use cases here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Case+Studies and talks on youtube such as https://www.youtube.com/watch?v=xq2LVY18GU8
and talks from recent CCC https://www.youtube.com/playlist?list=PLnIKk7GjgFlYcKNbhYSgWGbJ8nJ0g496V
> 3. The team could standardize on containers soon (especially for dev environments). I would tend to do a simple VM + Docker workflow. Is it standard?
Yes, this can be used. CloudStack IaaS platform doesn't care what you run inside the VM - you're free to decide on your stack and development choices.
> 4. I am all for simplicity of operation and maintenance. I think CloudStack could really shine here, right? I also think that Kubernetes should be avoided because of this. Thing is, Kubeflow is getting some attention and we might have to support it... Is CKS considered mature for production and a viable solution in this case?
CloudStack is boring and just works. Of course, I encourage you to setup a test env. and play with it, test and try all the requirements and see for yourself.
On container orchestration, you can setup and run whatever you'd like in VMs deployed via CloudStack or you can also use (a) CKS which is CloudStack's managed Kubernetes service offering, but we now also have (b) CAPC https://cluster-api-cloudstack.sigs.k8s.io<https://cluster-api-cloudstack.sigs.k8s.io/>. CAPC even works with EKS-A you can try. Both options are considered production and viable solution, but each have their pros and cons. If you're looking for more flexbility and not get tied into any IaaS platform then CAPC is the way to go. My suggestion is for you to test drive CloudStack.
You can try:
https://github.com/shapeblue/hackerbook/blob/main/1-user.md
You can follow our quickstart guide here: https://docs.cloudstack.apache.org/en/latest/quickinstallationguide/qig.html or if you want an Ubuntu-based guide here; https://rohityadav.cloud/blog/cloudstack-kvm/
If you've an Ubuntu desktop/laptop, you can even try mbx https://github.com/shapeblue/mbx to deploy CloudStack with one of KVM/XenServer/XCP-ng/VMware hypervisors (in a nested env, i.e. on a VM) with a community or 3rd party repo (see https://cloudstack.apache.org/downloads.html).
Regards.