19 posts • joined 11 Jun 2013
Re: Headcount growth reduces revenue?
It was the latter - we've changed the line to state AWS may be affecting overall profitability. Thanks for spotting our accidental conjunction of AWS hiring with revenue slowdown.
My view: this is what competition looks like!
You're right - I will read up on this and try to have a chinwag with HP about this issue then write it up. As I say in the article, I wish them the best in developing it - we've been waiting for a memristor for a long, long time. Thanks!
Re: Expected response from Microsoft minions Dell and The Register.
Actually, my opinion is that it's confusing HP is doing a completely new open source OS rather than trying to do this within the Linux community.
Re: Destroyed All Braincells Meh.
Hiya, You may be mistaken - a reader got in touch recently to tell me that IBM recently demoed a beta of Platform Symphony LSF using RAPL for power management, so I'm not sure Google are that far behind (or even behind at all, given that this work at Google was likely done some months/years ago). Also, the distributed controller scheme strikes me as novel as well.
Stay tuned for Haiku
Be very afraid
Re: doesn't seem very amazing
Hi, Nate - actually it is a full neural network so somewhat more than some scripts. Technical details are given in the detailed whitepaper linked to from within Google's blog post http://googleblog.blogspot.com/2014/05/better-data-centers-through-machine.html
Hope that helps!
Hello, unfortunately the article also waffles on about Jeff Hawkins, so Hawking/Hawking's/Hawkins all present but, as far as I can work out, correctly attributed.
Google is believed to run as many of its services it can off of a very small set of highly-integrated bits of software. For instance, it has a resource scheduler/cluster fiddler named Borg (new version: Mesos) that I've heard is in charge of the majority of Google jobs worldwide. I've even heard that Larry Page would like the company to ultimately use a single database across all of its services, but that has not happened yet (nor may ever be feasible). That said, we do know that it uses stuff like Spanner and F1 across a huge amount of services. So, when I say monolithic, I mean that the majority of Google's services are believed to draw upon a single mesh of services, so problems in one app may be indicative of a wider problem.
Much like the works of Stephen King, I found the first two thirds of it brilliant and the last third tepid and dull.
Re: Question on server types
Front-end Web & News Feed. Bit more info here (PDF) http://opencompute.org/summit/ [note - in this slide preso Haystack == Photo]
Nate - spot on
Hey Nate - you're right, Red Hat does use this exact tech in its OpenShift tech http://www.theregister.co.uk/2013/09/19/wtf_is_containerisation_and_why_are_red_hat_and_dotcloud_doing_it/
I'm currently reworking the model to include further info, and will figure out a way to expose it in a more useful manner either via a live sheet or a data dump at some point soon. Cheers!
Re: What kind of instances?
Hullo - these are all on-demand workloads, which I believe are meant to be about 60/70% of overall market, though YMMV.
It is stored in iCalendar. The file formats were not in the press release, but on the detailed FAQ page. https://support.google.com/accounts/answer/3024195?hl=en#calendar
I asked Oracle about this and Kurian said they are using it as a provisioning platform and have made numerous enhancements to it. I will be seeking more information throughout the show."
Quanta and other Asian ODMs started building custom servers for major clouds and then took marketshare from Dell DCS. Now HP is behind and having to involve in OCP and Moonshot to get back in with cut-down build to fit models, as is - I understand - Dell.
Soylent all the way
Mr Coatsworth - my daily diet consists of Soylent for breakfast, Soylent for lunch, and Soylent for dinner. Along with this, I've been having coffee every now and then, and on the weekends lashings of beer/spirits/dubious alcoholic concoctions.
Re: Crass speculation by article author
Mr Anderson, I believe OpenStack is several years behind Amazon in terms of the amount of features and services it can offer. Its core storage, compute, and network have rapidly matured, but it lacks the broad ecosystem of additional techs that defines Amazon. For proof, please refer to any list of OpenStack's techs and Amazon's techs.
A common complaint I hear from major public cloud users is that OpenStack is "AWS circa 2009". I also speak to companies that want to transition off of Amazon for a variety of reasons and many of them have the same problem - few other clouds have the broad set of capabilities AWS has. If they could move to OpenStack, they would, but it isn't ready. This, of course, will change - but it is not there yet.
I have covered OpenStack closely since its formation in mid-2010. When it is competitive, I'll be the first to say so. Thanks for reading, JC
Re: Different to everyone except VMware, Joyent...
Hi there. Neither VMware or Joyent sell both the hardware and software into ISPs - that is the unique part.
- +Analysis Microsoft: We're making ONE TRUE WINDOWS to rule us all
- Climate: 'An excuse for tax hikes', scientists 'don't know what they're talking about'
- Apple: We'll unleash OS X Yosemite beta on the MASSES July 24
- Pics It's Google HQ - the British one: Reg man snaps covert shots INSIDE London offices
- White? Male? You work in tech? Let us guess ... Twitter? We KNEW it!