Virtualization and HA, Scalability
-
@rojoloco said in Virtualization and HA, Scalability:
@dustinb3403 said in Virtualization and HA, Scalability:
Is there a question or are you attempting to define virtualization?
Sounds like a hell of a homework assignment.
Yeah, the scope is huge.
-
@kelsey said in Virtualization and HA, Scalability:
@rojoloco its one of the question on the assignment but i was struggling with just this one
We are struggling with it, too
-
The big answer, in the virtualization space today, is hyperconvergence. It is the only architecture for virtualization that is designed to address this issue rather than to address a different issue and then has to be adjusted to handle HA.
Nearly all HA deployments done today should be hyperconverged, rather than anything else.
-
@scottalanmiller at least you are trying to help people in my group help each other but not me
-
@kelsey said in Virtualization and HA, Scalability:
@scottalanmiller at least you are trying to help people in my group help each other but not me
Well they are your competition, I suppose
When teachers grade on a curve, it pits students against each other rather than encouraging learning.
-
Technically scalability and HA are completely different concepts. But in modern system design, they come and go together, making the answer often pretty easy. Old style scalability was at odds with HA (the inverted pyramid of doom model) so finding a way to do both was challenging (read: expensive.) But with hyperconvergence, the two are essentially merged and it isn't hard to find them almost always together today. So products that solve one, almost always solve the other.
-
@scottalanmiller thanks for helping
-
@kelsey said in Virtualization and HA, Scalability:
@scottalanmiller thanks for helping
Any time. We are always here and try very hard to be supportive and welcoming. We have mostly all been through the IT university process and we know how tough it is.
Glad to have you here joining us. -
Some vendors in hyperconvergence to keep an eye on would include @StarWind_Software / @KOOLER and @scale Computing, as well as HPE's Simplivity ( poking @HPEStorageGuy ) as they are leading products and companies and are all active here should you want to talk to any directly. There are lots of players getting into this market, like VMware, Dell, Nutanix, etc. and loads of confusing products. But those three are early leaders with a lot of knowledge and support.
-
Highly Available systems that are hyperconverged systems typically make the most sense... as Scott said, they also have the benefit of being very scalable. You can easily add more compute, storage, or other resources as needed, by adding more nodes or components to existing nodes.
-
@tim_g thanks
-
@scottalanmiller thanks for the mention. As was mentioned, Scale Computing makes scalable, highly available hyperconverged virtualization platform solutions. If there are any questions that we can answer about our products or HC concepts, we're here to assist. Sounds like a good class project and a chance to provide a lot of material for your class.
Is this written work or do you get to give a presentation?
-
@scale this is written work but we have done a presentation to
-
@kelsey said in Virtualization and HA, Scalability:
@scale this is written work but we have done a presentation to
If you needed presentation materials, I'm sure that we could find some for you.
-
@scale we have done the presentation that was about proxmox , VSphere and things like that, that was group work but the written work is on my own
-
@kelsey said in Virtualization and HA, Scalability:
@scale we have done the presentation that was about proxmox , VSphere and things like that, that was group work but the written work is on my own
Those are all interesting projects (we could have a long discussion about how Proxmox is a scam company, but we can save that for another time, but they get no love in technical communities as they run big scam review games to try to convince people that someone uses them), but they are just pieces of a puzzle when used on their own. None of them do the things that you need, but can be part of what you are looking at.
-
@scottalanmiller we had to use them for the presentation but that done i think anyway
-
@kelsey said in Virtualization and HA, Scalability:
@scottalanmiller we had to use them for the presentation but that done i think anyway
Was the presentation how they failed to fit the need? LOL
-
@scottalanmiller this was for the presentation -Each group will be required to deliver a presentation of about 30 minutes to the whole class, showing what they have installed including the major system components and salient features. The presentation should also include a critical evaluation of the applicability of the solution to the problems faced by the computing industry.
-
@kelsey said in Virtualization and HA, Scalability:
@dustinb3403 thats the question problems faced by organisations implementing highly available and scalable computing resources, evaluation of potential solutions, selection of a solution for the organisation chosen and discussion and critical evaluation of implementations with particular reference to the lab systems installed.
One of the best ways to tackle this is to present unique scenarios and argue for implementations.
This isn't the kind of thing that can be chosen generically. The workloads, budget, value, performance, scale - all matter, a lot. Some workloads, like Active Directory or databases, would require that the features than most would promote be disabled.
Hopefully the class is tackling the idea of high availability in layers - HA at this layer would ignore application failure, for example.
