=== CyberJacob is now known as CyberJacob|Away === mwhudson is now known as 5EXAAR2RG === 5EXAAR2RG is now known as mwhudson_ [03:51] jtv: one thing I forgot to say that we need to think about is to have a way indicate that particular nodes should use a particular bootresource [03:52] tags seems the obvious choice, but it would need hooking up [03:53] Shouldn't that be controlled by the settings we already have, plus a "label" field? [03:53] jtv: how would we currently direct a specific node to use a specific kernel? [03:54] once labels are implemented, I mean [03:55] Aren't arch/subarch/release/label/purpose already the main constraints for that though? [03:56] Although I guess tags make sense for HWE. === zchander_ is now known as zchander === Lord_Set2 is now known as Lord_Set [11:05] Greetings [11:09] Hi Lord_Set [11:11] How are you jtv? [11:11] Fine here, thanks. You? [11:12] Good. Ready for work to slow down some lol. [11:12] That's got to be good, the rate you've been going. [11:12] 20+ hour days for the past few weeks... [11:12] /o\ [11:14] Also have some pretty awesome new things in the works. Can you imagine 4800 cores, 1200 phsyical nodes, within 1 rack? [11:14] Nice rack. [11:14] That will be a good test for MAAS ;) [11:14] Yes it will. [11:15] We got the wild hair up our asses about building a scalable and very cost effective solution using ARM. [11:16] Only way you're going to get those numbers! [11:16] Using quad-core A9 or A11 boards with atleast 8gb DDR3 and 8gb internal flash. [11:16] Would be interesting to see where the bottlenecks are. If you have a large number of those racks, you're going to need a fast server. [11:16] Yeah [11:17] You might hit tftp bottlenecks, in which case... smaller clusters. [11:17] Well they would all be running 1g ethernet... currently we are already doing testing and using dual 10g on our cluster controller [11:17] *slobber* [11:18] I bet YouTube's pretty good on that. [11:18] We could easily do 2 10g dual port nics in a port channel [11:18] So have 40g possible for the cluster controller... I am sure that could handle it all [11:19] We've seen some slowness (ages ago though) with the tftp server; it may not be a network bottleneck. [11:19] Which tftp server package are you guys using for MAAS? I never really looked into it. [11:22] We got it for free with something else, but I need to remember what that was. [11:22] It's Twisted-based. [11:22] Haven't heard anything about slowness lately, so maybe that was just something that went away. [11:23] tx-tftp..? Lemme look. [11:24] Hmm [11:24] Yes, python-txtftp. [11:24] I have so ideas that you could could do to internally cluster or load balance TFTP to increase throughput... TFTP is a finicky and picky protocol. [11:25] err have some [11:28] I will shoot them to you privately in a bit [11:28] Let's hope we don't need to get too creative though! [11:28] Especially when it's third-party code. [11:30] True. I could have one of the developers on my team write a new custom built multi-thread and instance tftp server. [11:30] If we start to run into some issues... We will see soon as even outside of the ARM scenerio we will be hitting super high density between physical servers and VM. [11:31] The thought of getting that makes me a little giddy... But as always: remember Knuth's Law or suffer the consequences. :-) [11:31] Should be hitting 160-200 MAAS nodes per rack depending on the specific hardware used. This will be deployed within a month max. [11:33] Are you expecting a big jump to the big racks, or a steady ramp? [11:33] To the 160-200 node racks will be a big jump and move. [11:34] And to the 4K-node racks? [11:34] The ARM idea is a development idea and most likely 3-6 months off. [11:34] Ah OK. [11:35] We will have to have some hardware fabricated for it... a power and ethernet backbone for each rack case [11:35] Were you thinking to have an x86 cluster controller with ARM nodes? Or all ARM? [11:35] x86 most likely just for the power. [11:35] Yeah. [11:36] Also unsure on exactly what will be their best use. Possibly Hadoop or cheap web server nodes. [11:36] Also, Python is quite memory-hungry so with large numbers of objects, address space might become an issue if we're talking 32-bit ARM. [11:36] I don't *think* we're at the level yet where a few thousand nodes might be an issue, but the proof of the pudding is in the eating. [11:37] Well I would prefer to use 64bit Intel Alterra ARM chips. [11:39] Or something similar [11:39] Haven't looked at those. Are they available? [11:39] Pretty sure they are [11:39] Pretty awesome 14nm chips [11:40] You can get anything ARM produced super cheap in China lol [11:40] I'm sure you can! [11:41] Or Taiwan, I guess. [11:41] Is the Altera licensed as a design, or only sold as a chip? [11:42] How do you think companies like Minix can sell quad-core arm boxes with 4gb ram and 8gb flash, hdmi, optical audio, gigabit ethernet, 2x usb 2.0, 802.11n, a remote and cables and power for 100 dollars per... [11:42] That includes packaging and shipping [11:43] More or less the same chip that is in the Samsung Galaxy S4 [11:43] Yeah that world never ceases to amaze me. [11:44] ARM is for sure the future of high density computing === edu-afk_ is now known as edamato [17:34] hi bigjools, no luck even with syncdb on running the maas dev env [17:34] bigjools: now i'm seeing http://paste.ubuntu.com/7080407/ [17:35] bigjools: do you guys have a 'just-run-the-damn-thing' target in the makefile? === edamato is now known as edu-afk === zchander is now known as zchander_ [20:41] can you create a zone spanning multiple maas-clusters/subnets? Or do zones create groups within one cluster/subnet? === kevin is now known as Guest68425