search

cncf / cluster

🖥🖥🖥🖥CNCF Community Cluster
https://cncf.io/cluster
153 stars 42 forks source link

Request for access for Vitess Project (CNCF member project) #107

Closed dkhenry closed 5 years ago

dkhenry commented 5 years ago

Please fill out the details below to file a request for access to the CNCF Community Infrastructure Lab. Please note that access is targeted to people working on specific open source projects; this is not designed just to get your feet wet. The most important answer is the URL of the project you'll be working with. If you're looking to learn Kubernetes and related technologies, please try out Katacoda.

First and Last Name

Daniel Kozlowski

Email

koz@planetscale.com

Company/Organization

Planetscale

Job Title

Minister of Engineering

Project Title (i.e., summary of what do you want to do, not what is the name of the open source project you're working with)

Native Large Scale Vitess Testing

Briefly describe the project (i.e., what is the detail of what you're planning to do with these servers?)

Vitess has been run at massive scale by various companies, but there is currently no public large scale tests that have been run. Running on packet would be an excellent performance comparison to other workloads that have been used. Also vitess has first class support for multiple datacenters, but the open source implementation of failover is not as broadly tested as the proprietary versions

Is the code that you’re going to run 100% open source? If so, what is the URL or URLs where it is located? What is your association with that project?

Yes, github.com/vitessio/vitess, I am a maintainer

What kind of machines and how many do you expect to use (see: https://www.packet.com/bare-metal/)?

m2.xlarge c2.medium (maybe) c1.large.arm

What OS and networking are you planning to use (see: https://support.packet.com/kb/articles/supported-operating-systems)?

Linux

Any other relevant details we should know about?

dankohn commented 5 years ago

If it's just these 3 machines, this looks good. If you're looking for many instances of these 3 machines, we'll need to ask Packet to OK it.

+1

Dan Kohn dan@linuxfoundation.org Executive Director, Cloud Native Computing Foundation https://www.cncf.io +1-415-233-1000 https://www.dankohn.com

On Mon, Mar 4, 2019 at 6:48 PM Daniel Kozlowski notifications@github.com wrote:

Please fill out the details below to file a request for access to the CNCF Community Infrastructure Lab. Please note that access is targeted to people working on specific open source projects; this is not designed just to get your feet wet. The most important answer is the URL of the project you'll be working with. If you're looking to learn Kubernetes and related technologies, please try out Katacoda https://www.katacoda.com/courses/kubernetes/playground. First and Last Name

Daniel Kozlowski Email

koz@planetscale.com Company/Organization

Planetscale Job Title

Minister of Engineering Project Title (i.e., summary of what do you want to do, not what is the name of the open source project you're working with)

Native Large Scale Vitess Testing Briefly describe the project (i.e., what is the detail of what you're planning to do with these servers?)

Vitess has been run at massive scale by various companies, but there is currently no public large scale tests that have been run. Running on packet would be an excellent performance comparison to other workloads that have been used. Also vitess has first class support for multiple datacenters, but the open source implementation of failover is not as broadly tested as the proprietary versions Is the code that you’re going to run 100% open source? If so, what is the URL or URLs where it is located? What is your association with that project?

Yes, github.com/vitessio/vitess, I am a maintainer What kind of machines and how many do you expect to use (see: https://www.packet.com/bare-metal/)?

m2.xlarge c2.medium (maybe) c1.large.arm What OS and networking are you planning to use (see: https://support.packet.com/kb/articles/supported-operating-systems)?

Linux Any other relevant details we should know about?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/cncf/cluster/issues/107, or mute the thread https://github.com/notifications/unsubscribe-auth/AC8MBgrPGQZmFhxgQ8JLKJENnFlVjeJvks5vTbDKgaJpZM4bdcuc .

dkhenry commented 5 years ago

It would be many instances of those machines, normally I just count cores and I would like to run 1000 cores if I could, and I don't even know where to being with ARM. Also I don't know how many data centers I have access to, but we would like to run across multiple data centers.

dkhenry commented 5 years ago

Also fewer cores would be fine ( just let me know how many I can use ), and the test normally takes about 12 hours ( loading data, and then running TPC like workloads )

jacobsmith928 commented 5 years ago

Thanks @dkhenry for the details. How long would the scale test run?

dkhenry commented 5 years ago

About 12 hours or so, depending on how fast I can load a few TB of data

jacobsmith928 commented 5 years ago

Awesome @dkhenry could you try with m1.xlarge? This will allow me to give you the same config across four facilities. If that works, @taylorwaggoner can get you setup and you can grab:

EWR1 - 40 x m1.xlarge.x86 (960 cores) <- you can grab up to 60 of these if you want. SJC1 - 3 x m1.xlarge.x86 (72 cores) NRT1 - 10 x m1.xlarge.x86 (240 cores) AMS1 - 4 x m1.xlarge.x866 (96 cores)

Sort of imbalanced but hopefully it accomplishes what you need. Let me know any feedback.

dkhenry commented 5 years ago

That will be awesome. The unbalanced deployment is actually closer to a real world setup that completely balanced setups

taylorwaggoner commented 5 years ago

@dkhenry, I've created your project and invited you to it in Packet. Please let me know if there's anyone else you'd like added. Thanks!