kijiki

31 ms·

271.
by kijiki 11y ago
Thanks for the kind words for Cumulus! - nolan
272.
by kijiki 11y ago
See appendix A.
273.
by kijiki 11y ago
"Failure Oblivous Computing" https://www.doc.ic.ac.uk/~cristic/papers/fo-osdi-04.pdf
274.
by kijiki 12y ago
I took a tour of the ATLAS experiment last year, and the guide said that if the LHC was operating, the radiation where we were standing would be fatal in seconds.
275.
by kijiki 12y ago
"OpenBSD's OpenBGPD responsible for 30% of worldwide BGP communication" Ummm. Citation needed?
276.
by kijiki 12y ago
Lawyers primarily highlight risks. It is easy to ignore them if you're convinced "it couldn't happen here (tm)".
277.
by kijiki 12y ago
How did you get office space at Moffett Federal Airfield?
278.
by kijiki 12y ago
DAC is a common term for this. Direct Attached Copper. It is a SFP+ plug on each end (but without all the optical magic), connected with twin-ax cable, which is like coax but with 2 signal paths, one for each direction.
279.
by kijiki 12y ago
You are limited by the hardware, and what our code supports programming into it. The big advantages are reusing config management tools like puppet/chef/ansible/etc, and monitoring tools like collectd/graphite/nagio
280.
by kijiki 12y ago
Stay tuned for low-end 1G platforms running Cumulus Linux. - nolan
281.
by kijiki 12y ago
I'll take a look, thanks for the link. We don't do OpenFlow, but L3 and custom apps are our main focus. That said, often times the subset of things that can be done w/ OpenFlow in actual existing hardware are things we can do
282.
by kijiki 12y ago
Basically anything on here: http://cumulusnetworks.com/support/linux-hardware-compatibil... Some easy ways to buy are: https://bm-switch.com/ http://whiteboxswitch.com/ http:/&#x
283.
by kijiki 12y ago
To clarify, the only part that is closed is "switchd", which is a userspace program that watches the kernel data structures (route tables, neighbor tables, bridges, ports, vlans, etc) and programs the hardware to match. It links
284.
by kijiki 12y ago
Could you elaborate on 9p/virtio performance? I've had pretty good results booting qemu/kvm VMs with 9p root filesystems... UML used to have some host kernel patches that were a lot faster for intercepting syscalls. There w
285.
by kijiki 12y ago
10 mm wafers? Were they ever that small? I've never heard of less than 1 inch...
286.
by kijiki 12y ago
This is only true if the cause is pilot error. If the cause is improper maintenance, the responsible party is still alive. In the case of the NTSB, there is a strong culture of not penalizing errors unless they were criminal or egregious,
287.
by kijiki 12y ago
At least you didn't invest in Teledesic...
288.
by kijiki 12y ago
If you disable rtti and exceptions, the only piece that needs runtime support is global constructors. I've done it in the past with linker tricks to generate an ELF section with the list of ctor addresses. The asm that calls main (whi
289.
by kijiki 12y ago
We've been working with the kernel folks on trying to get support for forwarding ASICs into the kernel, but it is probably going to take a while. For now, ASIC programming is done from userspace by a daemon we call "switchd"
290.
by kijiki 12y ago
I'm the CTO of Cumulus, and I still sometimes use ifconfig. My fingers are just too programmed at this point. That said, ifconfig doesn't support easy addition of multiple IP addresses to a single interface. You have to create 1
291.
by kijiki 12y ago
Sometimes you gotta play the marketing game, for better or for worse. If you're interested in having a substantive technical discussion about Cumulus, drop me a line: nolan@cumulusnetworks.com
292.
by kijiki 12y ago
I wouldn't characterize us as "anti-SDN", instead I would say that Cumulus Linux is not an SDN solution by itself. We work closely with all the major SDN vendors for use-cases that are virtualization-centric or cloud-centric.
293.
by kijiki 12y ago
It is stock Linux (Debian based), but behind the scenes, we're programming a hardware forwarding ASIC to do the actual packet forwarding. This is why we run on white-box switches and not servers, see our HCL for details: http:/&
294.
by kijiki 12y ago
If you're willing to leave the cisco/Arista/etc cocoon, NIC costs are actually the big ticket items. http://whiteboxswitch.com/collections/10-gigabit-ethernet-sw... $5595 for the switch, $699 for the sof
295.
by kijiki 12y ago
If you're going down the path of VXLAN support in Docker, I'd love to talk. The company I founded built a Linux distribution for commodity hardware switches that can do VXLAN encap/decap in hardware at 2+ Tbit/sec. The
296.
by kijiki 12y ago
"just switching namespaces" isn't easy, since a packet (in the kernel represented by an SKB) has to have an interface it came in on. The main role an veth pair has is to move the packet between namespaces, and to provide a n
297.
by kijiki 12y ago
The grant itself requires board approval, so that isn't a huge roadblock. The bigger issue is how much leverage you have. If you're experienced and they need you, getting double-trigger acceleration shouldn't be an issue, a
298.
by kijiki 12y ago
The exact timings depend on the exact chips used, and a specific CPU could be connected to many different types. You have to query something (often a eeprom) to determine what sort of timings you need, and then program the memory controller
299.
by kijiki 12y ago
It is based on the google tracking cookies, not being logged in. I don't have a google account, but I still see the house #s on firefox, until I flush their cookies.
300.
by kijiki 13y ago
KVM supports nested virtualization, so you can run Xen inside a KVM VM. VMware does as well. Xen paravirt has always run inside VMs, even before nesting support. I used to run it in VMware all the time for development purposes.

More ›