jbeda

18 ms·

91.
by jbeda 12y ago
sshd is kind of a poor example. Kubernetes has an idea of a pod -- a group of containers that share an netns and have an IP. Reasons you might want a pod: * A thick client or client side proxy that follows the ambassador pattern for service
92.
Julia Ferraioli's account of casual sexism in tech(blog.juliaferraioli.com)
21 points by jbeda 12y ago|4 comments
93.
by jbeda 12y ago
This could run on top of CoreOS (I haven't tried yet though). If we look at a very high level sketch of what a dynamic cluster system look like ( https://speakerdeck.com/jbeda/containers-at-scale?slide=5 ) I think t
94.
by jbeda 12y ago
I can't give specifics but a lot of these are short lived. For example, if you launch a MapReduce it'll typically launch containers for each of the workers and then take them down when the MR is done. This also doesn't speak
95.
by jbeda 12y ago
Yeah -- I talked about it a couple of weeks ago at GlueCon. Also shared that we launch over 2 billion containers every week. My slides from the talk: https://speakerdeck.com/jbeda/containers-at-scale PDF: http:/&
96.
by jbeda 12y ago
We also announced cAdvisor in go. This is a container metrics monitoring component. https://github.com/google/cadvisor
97.
by jbeda 12y ago
We are particularly excited about Kubernetes. We are taking the ideas for how we do cluster management at Google and creating an open source project to manage Docker containers. https://github.com/GoogleCloudPlatform/k
98.
by jbeda 12y ago
Thanks -- the template was done by the Google Cloud Platform marketing guys. Working with the extra wide format can be a little tricky. Just having text ends up looking really stark so I find myself adding a lot more supporting graphics an
99.
by jbeda 12y ago
Here it is hosted on GCS: http://slides.eightypercent.net/GlueCon%202014%20-%20Contain...
100.
by jbeda 12y ago
My slides from the talk at GlueCon: https://speakerdeck.com/jbeda/containers-at-scale
101.
Google Buys Stackdriver for Google Cloud Platform(googlecloudplatform.blogspot.com)
3 points by jbeda 12y ago|0 comments
102.
by jbeda 12y ago
I'm not up on what the plans for that team are right now. I can say that it is an explicit goal to expose more and more of Google's infrastructure through cloud -- including how we route traffic globally.
103.
by jbeda 12y ago
We announced Cloud DNS at Cloud Platform Live on March 25th. It kind of got lost in the noise around the big price cuts.
104.
by jbeda 13y ago
That quote made a big impressions to me also. It was from a Q&A session at UW from late 2011. Here is a video: http://www.geekwire.com/2011/its-hamburger-insights-bill-gat...
105.
by jbeda 13y ago
I was on the Google Talk team at the time when we introduced this feature and I can confirm we had no knowledge of other products called 'Off The Record'.
106.
by jbeda 13y ago
You can see how we implemented this in the GCE API with our operations resource. We had earlier versions that required long HTTP calls and it was generally less reliable if something should happen during the call.
107.
by jbeda 13y ago
I'm the tech lead for GCE. I'm sorry to hear that App Engine didn't work out for the poster. Perhaps someone from that team may have some suggestions. In addition, I'm happy that he was able to stay with the Google Cloud Platform. With t
108.
Tours of Duty: The New Employer-Employee Compact(hbr.org)
1 points by jbeda 13y ago|0 comments
109.
by jbeda 13y ago
These days it is all about dynamic range. A dynamically lit scene is much more challenging. Similar photos from phones at the same location looked much too contrasty.
110.
by jbeda 13y ago
I don't agree with this. While you might not need a DSLR, you can easily do better than the camera in your phone. I recently completed a family trip and left my Canon 5D mark III at home. Instead I took the pocketable Sony RX100. It created
111.
MinuteSort world record exceeded on Google Compute Engine(mapr.com)
1 points by jbeda 14y ago|0 comments
112.
by jbeda 14y ago
10 msec * c = ~3000 km or ~1860 miles. Of course light doesn't travel that fast through glass and there is latency at the hardware on either end. ;)
113.
by jbeda 14y ago
I was there when this was happening and realized exactly what was going on at the time. Still, it was very very hard not to respond with a 'Stop replying all' type message. There is something about these things that makes it hard to just
114.
Google Cloud Platform: new features, lower price, extending European datacenters(googledevelopers.blogspot.com)
9 points by jbeda 14y ago|0 comments
115.
by jbeda 14y ago
Yes -- the Capital One cards are well known for not charging a foreign transaction fee. I keep one just for international travel.
116.
by jbeda 14y ago
I see this type of thing come up all the time when monitoring complex production systems. Say you have 10 servers in each of 3 datacenters and you are looking at request latency. Averaging all 30 servers is very different from averaging to
117.
by jbeda 14y ago
We went back and forth on naming the machine types. I'm sure you can imagine those discussions. In the end we opted for naming them based on the number of virtual CPUs from inside the VM. This is easier to remember than arbitrary sizes (
118.
by jbeda 14y ago
I'd love to hear what people think the standard nomenclature here is. I've been arguing with our PM as to call these things virtual cores or virtual CPUs. Our API uses guestCpus. This stuff is very confusing and I'd love to find the right
119.
by jbeda 14y ago
I'd love to make our docs more clear on this. If you have a pointer to what is confusing we'll fix it up. To be clear, for every machine type, we are offering a HT per virtual CPU. So that means that an n1-standard-8 instance gets 4 physi
120.
by jbeda 14y ago
Sign up and let us know what you are looking to do. We are trickling out invites but being careful that we can deliver a great experience (support wise and otherwise) to everyone.

More ›